A vehicle-road-load integrated heavy-duty vehicle dynamics modeling method

Abstract

The present invention is a vehicle-road integrated heavy-duty vehicle dynamics modeling method. It includes the following steps: obtaining the road slope, road adhesion coefficient and road roughness, estimating the cargo center of mass; constructing a "cargo-whole vehicle" center of mass dynamics reconstruction matrix, and establishing a vehicle center of mass offset equation; based on the vehicle center of mass offset equation, combined with the mechanical characteristics of the transport vehicle when the cargo volume changes under different axle number configurations, derive the vertical force balance equation of each axle; analyze the coupling relationship between road roughness and wheel vertical force, derive the wheel-ground dynamics equation, construct a "wheel-ground-whole vehicle" dynamics reconstruction matrix, and design the transfer equation of wheel-ground contact dynamics; combine the "cargo-whole vehicle" center of mass dynamics reconstruction matrix and the "wheel-ground-whole vehicle" dynamics reconstruction matrix to achieve the mapping and transfer of wheel-ground contact dynamics to the whole vehicle dynamics state, and construct a vehicle-road integrated heavy-duty vehicle dynamics model. The present invention establishes a universal heavy-duty vehicle dynamics modeling method.

Description

Technical Field

[0001] The invention belongs to the field of vehicle dynamics modeling, and in particular relates to a vehicle-road-load integrated heavy-duty vehicle dynamics modeling method. Background Art

[0002] As the core carrier of modern logistics and transportation, the dynamic characteristics of heavy-duty vehicles are strongly coupled by vehicle configuration, road environment and cargo status, showing significant complexity and uncertainty. On the one hand, heavy-duty vehicles are diverse in models, including tractors, dump trucks, tank trucks and other configurations. Differences in wheelbase distribution, driving mode and other factors directly affect the inertia distribution and dynamic response of the vehicle. On the other hand, complex road environments (such as slope, road roughness, and dynamic changes in road adhesion coefficient) and the diversity of load states (involving different weights and distributions) are intertwined, further amplifying the nonlinear and multimodal characteristics of dynamic behavior. Traditional modeling methods, whether physically driven or data driven, face significant challenges.

[0003] 1. Physical drive method:

[0004] Physically driven methods are usually based on simplified assumptions, such as treating cargo as a rigid body with fixed mass, ignoring key details such as center of mass offset and dynamic changes in moment of inertia. At the same time, road models often use static friction coefficients, which cannot accurately capture real-time changes such as the sharp drop in friction coefficient on icy and snowy roads or friction performance degradation caused by long downhill braking. In addition, these methods often focus on the unidirectional effect of road excitation on vehicle vibration, while ignoring the role of the bidirectional dynamic feedback mechanism between the vehicle and the road in the vehicle's dynamic performance.

[0005] 2. Data-driven approach:

[0006] Data-driven methods, especially deep learning models, can fit interaction rules through a large amount of specific working condition data training, but their black box characteristics limit the interpretability of the model and make it difficult to generalize to different types of vehicle models or unforeseen load conditions. In the face of heterogeneous vehicle models or extreme load conditions, the prediction accuracy of the model will drop significantly. At the same time, the static training mechanism makes it difficult for the model to adapt to changes in dynamic working condition parameters, and model drift problems are prone to occur.

[0007] In the process of model building, there is also an inherent contradiction between the accuracy of dynamic modeling and real-time performance. The high-precision multi-body dynamics model has a sharp increase in computational complexity due to the significant increase in the degree of freedom, making it difficult to meet the real-time requirements of the vehicle system. Although the simplified model improves the computational efficiency, the excessive linearization processing makes it ineffective under extreme conditions such as emergency obstacle avoidance. These limitations jointly restrict the performance of traditional methods in high-dynamic and strong interference environments, becoming a key technical problem hindering the development of heavy-duty vehicle intelligence. Summary of the invention

[0008] The purpose of the present invention is to provide a vehicle-road-load integrated heavy-duty vehicle dynamics modeling method to solve the problem that the existing heavy-duty vehicle dynamics model is subject to the strong coupling of vehicle configuration, road environment and load state, presenting significant complexity and uncertainty.

[0009] The technical solution to achieve the purpose of the present invention is: a vehicle-road-load integrated heavy-duty vehicle dynamics modeling method, comprising the following steps:

[0010] Step (1): Obtain the road slope, road adhesion coefficient and road roughness, and estimate the center of mass of the cargo;

[0011] Step (2): Construct the "cargo-vehicle" center of mass dynamics reconstruction matrix and establish the vehicle center of mass offset equation;

[0012] Step (3): Based on the vehicle center of mass offset equation and combined with the mechanical characteristics of the transport vehicle with different axle configurations and the change in the amount of cargo loaded, derive the vertical force balance equation for each axle;

[0013] Step (4): Analyze the coupling relationship between road surface roughness and wheel vertical force, derive the wheel-ground dynamics equation, construct the "wheel-ground-whole vehicle" dynamics reconstruction matrix, and design the transfer equation of wheel-ground contact dynamics;

[0014] Step (5): Combine the “cargo-vehicle” center of mass dynamics reconstruction matrix and the “wheel-ground-vehicle” dynamics reconstruction matrix to achieve the mapping and transfer of wheel-ground contact dynamics to the vehicle dynamics state, and construct a vehicle-road-load integrated heavy-duty vehicle dynamics model.

[0015] Furthermore, step (1) "obtaining the road surface slope, road surface adhesion coefficient and road surface roughness" is specifically as follows:

[0016] Step (11): classifying the road surface according to the road adhesion coefficient;

[0017] Step (12): Acquire an image, and annotate the image according to the road adhesion coefficient classification and the road roughness classification based on power spectral density (PSD) defined in step (11), and construct a training data set;

[0018] Step (13): Construct a Kalman filter model and train the Kalman filter model based on the YOLOv5 deep learning algorithm;

[0019] Step (14): Acquire the road surface image in real time, input the trained Kalman filter model, and obtain the road surface adhesion coefficient μ and road surface roughness.

[0020] Furthermore, step (1) “estimating the center of mass of the goods” is as follows:

[0021] The pressure sensor obtains the mass of the goods, obtains the shape and volume of the goods based on the image, and estimates the center of mass coordinates of the goods.

[0022] Furthermore, step (2) specifically includes the following steps:

[0023] Step (21): Set the initial vehicle center of mass equation:

[0024] ,

[0025] in, is the unladen mass of the vehicle, is the coordinate of the vehicle center of mass, For the quality of goods, is the coordinate of the cargo center of mass, is the coordinate of the center of mass of the vehicle;

[0026] Step (22): Construct the “cargo-vehicle” center of mass dynamics reconstruction matrix :

[0027] ,

[0028] in, is the identity matrix,

[0029] Step (23): Reconstruct the matrix based on the center of mass dynamics of the cargo-vehicle , mapping the change of cargo center of mass to the displacement of vehicle center of mass, and obtaining the displacement equation of vehicle center of mass:

[0030] ,

[0031] in, is the center of mass offset of the vehicle.

[0032] Furthermore, the “vertical force balance equation of each axle” in step (3) is specifically:

[0033] ,

[0034] in, is the number of vehicle axles, is the number of wheels on each axle, is the wheelbase of each axle, , is the wheel track, is the horizontal coordinate of the centroid, is the centroid height, is the longitudinal acceleration, is the acceleration due to gravity, is the lateral acceleration, It represents the vertical force on the jth wheel of the i-th wheel axle of the vehicle;

[0035] Convert the vertical force balance equation of each axle into matrix form:

[0036] ,

[0037] in, , , is the Kronecker product, and the wheel vertical force distribution is solved by the pseudo-inverse matrix:

[0038] ,

[0039] in, , .

[0040] Further, step (4):

[0041] Step (41): The coupling relationship between the vertical force and the road surface parameters is as follows:

[0042] ,

[0043] in, For road roughness, is the tire displacement, , are the tire roll coefficient and stiffness coefficient, is the road surface elevation generated by the power spectral density (PSD);

[0044] Step (42): The longitudinal acceleration calculation formula of the wheel-ground dynamics equation is:

[0045] ,

[0046] in, Driving force for each wheel Braking force the sum of , , for The driving torque, transmission efficiency and driving gear radius of the wheel, is the tire radius, for The brake pressure of the wheel, is the braking efficiency factor, is the air resistance, which is determined by the air density and windward area , drag coefficient and longitudinal speed The calculation formula is as follows:

[0047] ,

[0048] Wheel longitudinal and lateral force model of wheel-terrain dynamics equation:

[0049] ,

[0050] ,

[0051] in, is the slip rate, is the slip angle, is the road adhesion coefficient, function Defined as:

[0052] ,

[0053] ,

[0054] in, , are the equivalent longitudinal adhesion coefficient and lateral adhesion coefficient, respectively. , , , , , are the longitudinal and transverse stiffness factors, curve shape factors and peak factors;

[0055] Step (43): Construct the “wheel-ground-vehicle” dynamic reconstruction matrix as follows:

[0056] ,

[0057] in, , is the tire position vector of the i-th axis;

[0058] Step (44): Design the transfer equation for wheel-ground contact dynamics:

[0059] ,

[0060] in, is the tire force of each wheel, is the external force vector acting on the vehicle, is the vehicle angular velocity, is the vehicle acceleration vector, is the moment of inertia,

[0061] Expanded into the Newton-Euler equation:

[0062] ,

[0063] in, is the vehicle longitudinal velocity, is the vehicle lateral speed, and are the moment of inertia and angular velocity of the vehicle around the z-axis, respectively.

[0064] Furthermore, the vehicle-road integrated heavy-duty vehicle dynamics model constructed in step (5) is as follows:

[0065] ,

[0066] in, is a callable parameter set, is the vehicle state vector, is the inertia matrix, is the Coriolis force matrix, is the tire force projection matrix, is the tire force vector, is the external force vector, updated when the cargo load changes and , recalculate the vertical force on the wheel ; Update when road conditions change , In ,Adjustment The power spectral density parameter and .

[0067] Compared with the prior art, the present invention has the following significant advantages:

[0068] This application solves the problem of model generalization under the conditions of complex configuration of heavy vehicles, random road disturbances and variable loads, realizes the transfer and merging of wheel-ground contact dynamics and cargo specificity to the dynamic state of the whole vehicle, and changes the vehicle configuration parameters and actuator configuration by adjusting matrix elements, thereby forming an integrated heavy-duty vehicle dynamics general modeling framework that includes random road environment, vehicle configuration change characteristics and specific state of on-board cargo.

[0069] The present invention solves the problem caused by the diversity of cargo states (different weights and distributions) by constructing a "cargo-whole vehicle" center of mass dynamics reconstruction matrix; the "cargo-whole vehicle" center of mass dynamics reconstruction matrix effectively solves the influence of cargo center of mass offset and dynamic changes in moment of inertia on the vehicle state, so that the vehicle dynamics model can better cope with changes in cargo states.

[0070] The "wheel-ground-whole vehicle" dynamic reconstruction matrix of the present invention constructs a model of the role of the two-way dynamic feedback mechanism between the vehicle and the road surface in the vehicle's dynamic performance; in the case of different vehicle models and numbers of axles, the vehicle dynamics formula can be directly derived by inputting vehicle parameters to form a universal vehicle dynamics model; this solves the problem of poor universality of previous models, and there is no need to repeatedly model different vehicle models. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 This is a flow chart of the vehicle-road integrated heavy-duty vehicle dynamics modeling method of the present invention. DETAILED DESCRIPTION

[0072] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0073] The present application provides a vehicle-road-load integrated heavy-duty vehicle dynamics modeling method to solve the problem that the existing heavy-duty vehicle dynamics model is subject to the strong coupling of vehicle configuration, road environment and load state, presenting significant complexity and uncertainty. Figure 1 This is a general model diagram of the vehicle-road integrated heavy-duty vehicle dynamics of this application, such as Figure 1 As shown in FIG, the vehicle-road integrated heavy-duty vehicle dynamics modeling method includes the following steps:

[0074] S1: Obtain the road slope, road adhesion coefficient and road roughness to estimate the cargo quality;

[0075] S2: Construct the "cargo-vehicle" center of mass dynamics reconstruction matrix and establish the vehicle center of mass offset equation;

[0076] S3: Based on the vehicle mass center offset equation in S2, combined with the mechanical characteristics of the transport vehicle when the load changes under different axle configurations, the vertical force balance equation of each axle is derived;

[0077] S4: Analyze the coupling relationship between road roughness and wheel vertical force, derive wheel-ground dynamic equation, construct "wheel-ground-whole vehicle" dynamic reconstruction matrix, and design the transfer equation of wheel-ground contact dynamics;

[0078] S5: Combine the "cargo-vehicle" center of mass dynamics reconstruction matrix in S2 and the "wheel-ground-vehicle" dynamics reconstruction matrix in S4 to achieve the mapping and transfer of wheel-ground contact dynamics to vehicle dynamics state, and construct a "vehicle-road-load" integrated heavy-duty vehicle dynamics model;

[0079] In step S1, the road surface slope, road surface adhesion coefficient and road surface roughness are obtained, specifically:

[0080] S11: Classify the road surface according to the road adhesion coefficient;

[0081] S12: Acquire an image, and annotate the image according to the road adhesion coefficient classification defined in step S11 and the road roughness classification based on power spectral density PSD, to construct a training data set;

[0082] S13: Build a Kalman filter model and train the Kalman filter model based on the YOLOv5 deep learning algorithm;

[0083] S14: Acquire the road surface image in real time, input the trained Kalman filter model, and obtain the road surface adhesion coefficient μ and road surface roughness.

[0084] S15: The pressure sensor obtains the mass of the goods, obtains the shape and volume of the goods according to the image, and estimates the center of mass coordinates of the goods.

[0085] In step S2, a "cargo-vehicle" center of mass dynamics reconstruction matrix is ​​constructed, the law of the change of the cargo center of mass and the change of the vehicle's overall center of mass is analyzed, and the center of mass offset equation of the vehicle is designed, including the following steps:

[0086] S21: Set the initial vehicle center of mass equation:

[0087] (1)

[0088] in, is the unladen mass of the vehicle, is the coordinate of the vehicle center of mass, For the quality of goods, is the coordinate of the cargo center of mass, is the coordinate of the center of mass of the vehicle.

[0089] S22: Constructing the “cargo-vehicle” center of mass dynamics reconstruction matrix , mapping the change in cargo volume to the offset of the vehicle's center of mass:

[0090] (2)

[0091] in, is the identity matrix.

[0092] S23: Reconstruction matrix based on the center of mass dynamics of "cargo-vehicle" , mapping the change of cargo center of mass to the displacement of vehicle center of mass, and obtaining the displacement equation of vehicle center of mass:

[0093] (3)

[0094] in, is the center of mass offset of the vehicle.

[0095] In step S3, the vertical force balance equation is obtained by combining the center of mass offset equation in S2. The vertical force balance equation of each wheel is expressed as:

[0096] (4)

[0097] in, is the number of vehicle axles, is the number of wheels on each axle, is the wheelbase of each axle, , is the wheel track, is the horizontal coordinate of the centroid, is the centroid height, is the longitudinal acceleration, is the acceleration due to gravity, is the lateral acceleration, Indicates Axle wheel axle The vertical force on each wheel.

[0098] The matrix form of the vertical force balance equation is:

[0099] (5)

[0100] in, , , is the Kronecker product, and the wheel vertical force is solved by the pseudo-inverse matrix:

[0101] (6)

[0102] in, , .

[0103] In step S4, the influence of road surface parameters on wheel-ground dynamics is analyzed, the wheel-ground dynamics equation is derived, the "wheel-ground-whole vehicle" dynamics reconstruction matrix is ​​constructed, and the wheel-ground contact dynamics transfer equation is designed;

[0104] S41: The coupling relationship between road roughness and wheel vertical force is as follows:

[0105] (7)

[0106] in, For road roughness, is the tire displacement, , are the tire roll coefficient and stiffness coefficient, is the road elevation generated by the power spectral density (PSD).

[0107] S42: The calculation formula of longitudinal acceleration in wheel-terrain dynamics equation is:

[0108] (8)

[0109] in Driving force for each wheel Braking force The sum of , , for The driving torque, transmission efficiency and driving gear radius of the wheel, is the tire radius, for The brake pressure of the wheel, is the braking efficiency factor, is the air resistance, which is determined by the air density and windward area , drag coefficient and longitudinal speed The calculation formula is as follows:

[0110] (9)

[0111] The wheel longitudinal and lateral force model of the wheel-terrain dynamics equation is:

[0112] (10)

[0113] (11)

[0114] in, is the slip rate, is the slip angle, is the road adhesion coefficient, function Defined as:

[0115] (12)

[0116] (13)

[0117] in, , are the equivalent longitudinal adhesion coefficient and lateral adhesion coefficient, respectively. , , , , , are the longitudinal and transverse stiffness factors, curve shape factor and peak factor.

[0118] S43: Constructing the “wheel-ground-entire vehicle” dynamic reconstruction matrix as follows:

[0119] (14)

[0120] in, , is the tire position vector of the i-th axis.

[0121] S44: Design transfer equations for wheel-ground contact dynamics:

[0122] (15)

[0123] in, is the tire force of each wheel, is the external force vector acting on the vehicle, is the vehicle angular velocity, is the vehicle acceleration vector, is the moment of inertia,

[0124] Expanded into the Newton-Euler equation:

[0125] (16)

[0126] in, is the vehicle longitudinal velocity, is the vehicle lateral speed, and are the moment of inertia and angular velocity of the vehicle around the z-axis, respectively.

[0127] In step S5, the callable parameters in the center of mass offset equation in S2, the vertical force balance equation in S3, and the wheel tire force equation in S4 are proposed and integrated into a general heavy-duty vehicle dynamics model of "vehicle-road-load" integration after block division as follows:

[0128] (17)

[0129] in, is a callable parameter set, is the vehicle state vector, is the inertia matrix, is the Coriolis force matrix, is the tire force projection matrix, is the tire force vector, is the external force vector (air resistance, gravity), updated when the cargo load changes and , recalculate the vertical force on the wheel ; Update when road conditions change , In ,Adjustment The power spectral density parameter and .

[0130] The present invention discloses a general method for dynamic modeling of heavy-duty vehicles with vehicle-road integration, which breaks through the limitations of traditional heavy-duty vehicle dynamic modeling methods of "special vehicles for special purposes and fixed scenarios", and is suitable for solving the problem of generalization of models under conditions of complex configurations of heavy-duty vehicles, random road disturbances and variable loads. Aiming at the problem of non-universality of the modeling framework caused by complex factors such as time-varying road information, variable vehicle configurations, and variable loads, the present invention combines visual information with time series prediction technology to reveal the mechanism of action between parameters such as road slope, adhesion coefficient, and unevenness and tire mechanical response, and constructs "wheel-land-whole vehicle" and "cargo-whole vehicle" dynamic reconstruction matrices, mapping the wheel-land contact dynamic characteristics and cargo distribution states and integrating them into the whole vehicle dynamics model. By dynamically adjusting the matrix parameters, the optimal matching of vehicle configuration parameters and actuator configuration is achieved, and finally a generalized heavy-duty vehicle dynamic modeling method is established that integrates random road environment excitation, dynamic adjustment capability of vehicle configuration, and temporal and spatial distribution characteristics of load.

[0131] Example

[0132] Taking a three-axle transport vehicle as an example, the road adhesion coefficient is 0.6, the road roughness is grade A, and the cargo mass is obtained through visual information level time series prediction. , the centroid position ; Substitute the acquired road surface and vehicle parameters into S2 to update the "cargo-vehicle" center of mass dynamics reconstruction matrix , find the center of mass offset of the vehicle , update the center of mass coordinates of the vehicle , the vertical force balance equation of the vehicle is calculated by the center of mass position of the vehicle:

[0133] (18)

[0134] Get the matrix A= ,b= (19)

[0135] Solving the Pseudo-Inverse Matrix , from S3 Find the vertical force;

[0136] Reconstruction of the transfer matrix of the wheel-ground-vehicle dynamics in S43 :

[0137] (20)

[0138] The updated vehicle dynamics equations can be obtained as follows:

[0139] (twenty one)

[0140] The above are exemplary embodiments of the present application, and the protection scope of the present application is defined by the claims and their equivalents. Those skilled in the art should understand that the present application is not limited by the above examples, and the above examples and descriptions are only for illustrating the principles of the present invention. The present invention may have various changes and improvements without departing from the spirit and scope of the present invention, and these changes and improvements fall within the scope of the present invention to be protected. The protection scope of the present invention is defined by the attached claims and their equivalents.

Claims

1. A vehicle-road integrated heavy-duty vehicle dynamics modeling method, characterized in that: The steps include: Step (1): Obtain the road slope, road adhesion coefficient and road roughness, and estimate the center of mass of the cargo; Step (2): Construct the "cargo-vehicle" center of mass dynamics reconstruction matrix and establish the vehicle center of mass offset equation; Step (3): Based on the vehicle center of mass offset equation and combined with the mechanical characteristics of the transport vehicle with different axle configurations and the change in the amount of cargo loaded, the vertical force balance equation of each axle is derived; Step (4): Analyze the coupling relationship between road surface roughness and wheel vertical force, derive the wheel-ground dynamics equation, construct the "wheel-ground-whole vehicle" dynamics reconstruction matrix, and design the transfer equation of wheel-ground contact dynamics; Step (5): Combine the "cargo-vehicle" center of mass dynamics reconstruction matrix and the "wheel-ground-vehicle" dynamics reconstruction matrix to achieve the mapping and transfer of wheel-ground contact dynamics to the vehicle dynamics state, and construct a vehicle-road-load integrated heavy-duty vehicle dynamics model; Step (2) specifically includes the following steps: Step (21): Set the initial vehicle center of mass equation: Among them, m v is the unloaded mass of the vehicle, r v =[x v ,y v ,z v ] T is the coordinate of the vehicle center of mass, m c is the mass of goods, r c =[x c ,y c ,z c ] T is the coordinate of the cargo center of mass, r g is the coordinate of the center of mass of the vehicle; Step (22): Construct the "cargo-vehicle" center of mass dynamics reconstruction matrix C∈R 3×3 : Among them, I 3×3 is the identity matrix, Step (23): Based on the "cargo-vehicle" center of mass dynamics reconstruction matrix C, the change of the cargo center of mass is mapped to the displacement of the center of mass of the vehicle, and the center of mass displacement equation of the vehicle is obtained: Δr g =C(m c )·(r c -r v ), Among them, Δr g is the center of mass offset of the vehicle; The vehicle-road integrated heavy-duty vehicle dynamics model constructed in step (5) is as follows: Where p = [m c ,r c ,μ,L,B,…] T is a callable parameter set, x is the vehicle state vector, M(p) is the inertia matrix, C(p) is the Coriolis force matrix, T(p) is the tire force projection matrix, F tire (p) is the tire force vector, F ext (p) is the external force vector, and C(m c ) and r g , recalculate the wheel vertical force F z ; When the road conditions change, when the road adhesion coefficient changes, update μ x D x and μ y D y When the road surface roughness changes, adjust the road surface elevation z generated by the power spectrum density r .

2. The method according to claim 1, characterized in that Step (1) "obtaining the road surface slope, road surface adhesion coefficient and road surface roughness" is specifically as follows: Step (11): classifying the road surface according to the road adhesion coefficient; Step (12): acquiring an image, and annotating the image according to the road adhesion coefficient classification defined in step (11) and the road roughness classification based on power spectral density PSD, to construct a training data set; Step (13): construct a Kalman filter model and train the Kalman filter model based on the YOLOv5 deep learning algorithm; Step (14): Acquire the road surface image in real time, input the trained Kalman filter model, and obtain the road surface adhesion coefficient μ and road surface roughness.

3. The method according to claim 2, characterized in that Step (1) "Estimated Cargo Centroid" is as follows: The mass of the goods is obtained through the pressure sensor, the shape and volume of the goods are obtained based on the image, and the center of mass coordinates of the goods are estimated.

4. The method according to claim 3, characterized in that The "vertical force balance equation of each axle" in step (3) is specifically: Where n is the number of axles of the vehicle, k is the number of wheels on each axle, and L i is the wheelbase of each axle, i = 1, 2, ..., n, B is the wheel track, (x g ,y g ) is the horizontal coordinate of the center of mass, h is the height of the center of mass, a x is the longitudinal acceleration, g is the acceleration due to gravity, a y is the lateral acceleration, F zij It represents the vertical force on the jth wheel of the i-th wheel axle of the vehicle; Convert the vertical force balance equation of each axle into matrix form: Among them, F z =[F z11 ,…,F z1k ,F z21 ,…,F znk ] T ,L=[L1,L2,…,L n ] T , is the Kronecker product, is a 2n-dimensional unit vector, and the wheel vertical force distribution is solved by the pseudo-inverse matrix: F z =A + b,A + =(A T A) -1 A T , in, 5. The method according to claim 4, characterized in that Step (4): Step (41): The coupling relationship between road surface roughness and wheel vertical force is as follows: Among them, z r is the road elevation generated by the power spectral density PSD, F zr is the wheel vertical force based on road roughness, z tire is the tire displacement, k tire 、c tire are the tire roll coefficient and stiffness coefficient, For z r The derivative of For z tire The derivative of Step (42): The longitudinal acceleration calculation formula of the wheel-ground dynamics equation is: Among them, ∑F xi is the driving force F of each wheel idrive Braking force F ibrake The sum of T ei ,η,r gear is the driving torque, transmission efficiency and driving gear radius of wheel i, r tire is the tire radius, P bi is the braking pressure of wheel i, k b is the braking efficiency factor, F drag is the air resistance, which is determined by the air density ρ and the frontal area A t , drag coefficient C d and longitudinal speed v x The calculation formula is as follows: Wheel longitudinal and lateral force model of wheel-terrain dynamics equation: F x =F z m x (l,a,m), F y =F z m y (l,a,m), Among them, F x is the wheel longitudinal force, F y is the wheel lateral force, λ is the slip rate, α is the sideslip angle, μ is the road adhesion coefficient, and the function μ x (λ,α,μ) and μ y (λ, α, μ) are defined as: μ x =D x sin(C x silver(B x λ-E x (B x λ-arctane(B x λ)))), μ y =D y sin(C y silver(B y α-E y (B y α-arctan(α))), Among them, μ x , μ y are the equivalent longitudinal adhesion coefficient and lateral adhesion coefficient, B x is the longitudinal stiffness factor, B y is the lateral stiffness factor, C x is the longitudinal curve shape factor, C y is the transverse curve shape factor, D x is the longitudinal peak factor, D y is the transverse peak factor, E x is the longitudinal curvature factor, E y is the lateral curvature factor; Step (43): Construct the "wheel-ground-vehicle" dynamic reconstruction matrix T∈R 6×3n as follows: in, r i =[x i ,y i ,z i ] T is the tire position vector of the i-th axis; Step (44): Design the transfer equation for wheel-ground contact dynamics: Among them, F tire =[F x1 ,F y1 ,F z1 ,…,F xn ,F yn ,F zn ] T is the tire force of each wheel, F ext is the external force vector acting on the vehicle, and the angular velocity matrix of the vehicle around each axis ω z is the angular velocity of the vehicle around the z-axis, and the moment of inertia matrix around each axis I x is the moment of inertia about the x-axis, I y is the moment of inertia about the y-axis, I z is the moment of inertia about the z-axis, is the vehicle acceleration matrix, Expanded into the Newton-Euler equation: Among them, v x is the longitudinal velocity of the vehicle, v y is the lateral speed of the vehicle.

Images