Closed-loop automobile crosswind stability dynamic evaluation system based on MPC control algorithm

Abstract

The invention discloses a closed-loop automobile crosswind stability dynamic evaluation system based on an MPC control algorithm, and belongs to the technical field of intelligent driving. According to the system, a CFD model unit obtains pneumatic six-component parameters according to a first motion state parameter and a lateral wind environment parameter of a tested automobile at a first moment; the MPC active control unit predicts the state quantity and the control quantity of the tested automobile in a plurality of time steps according to the first motion state parameter on the basis of an MPC algorithm by taking the minimum control cost as a target so as to determine a front wheel steering angle parameter corresponding to the optimal control quantity; and the MBD model unit updates the dynamic parameters of the tested automobile according to the pneumatic six-component parameters and the front wheel steering angle parameters to obtain the dynamic response parameters of the tested automobile at the second moment, and evaluates the crosswind stability of the tested automobile according to the dynamic response parameters. The system is high in reduction degree of actual conditions, and the reliability of test results and the success rate of calculation are improved.

Description

Technical Field

[0001] The present application belongs to the field of intelligent driving technology, and in particular relates to a closed-loop automobile sidewind stability dynamic evaluation system based on an MPC control algorithm. Background Art

[0002] Automobile crosswind stability refers to the ability of a vehicle to maintain its driving direction and stability when subjected to crosswind. This performance is particularly important for high-speed vehicles. Strong crosswinds can cause the vehicle to move laterally in a manner inconsistent with the driver's wishes, which may cause the vehicle to deviate from its original driving route, or even cause dangerous driving situations such as skidding or rollover. Therefore, research on automobile crosswind stability has become a key way to improve vehicle safety performance.

[0003] Most of the current numerical simulation studies on the crosswind stability of automobiles in the industry need to be carried out on a high-computing power computing platform, and it is necessary to select an appropriate turbulence model to ensure the efficiency and accuracy of the solution. However, the existing methods are still insufficient in restoring the actual driving conditions, and the accuracy of the test results needs to be improved. First of all, the air flow around the car becomes extremely complex under the action of the crosswind, especially under high wind speeds and drastic gusts. This complexity is further exacerbated. Existing simulation methods, such as the Lattice Boltzmann Method (LBM), etc., although they can simulate this flow to a certain extent, their applicable velocity discrete models and boundary condition methods are still very limited, and it is difficult to completely restore the complex airflow changes in real scenes.

[0004] Secondly, existing simulation tests often ignore the complex coupling relationship between the driver's active correction behavior or the electronic stability system of the vehicle and the vehicle's driving posture and crosswind aerodynamic characteristics. In fact, changes in the vehicle's driving posture will in turn affect the crosswind aerodynamic characteristics, and active control of the direction will affect the vehicle's driving posture, the latter of which is not fully reflected in the current simulation model.

[0005] Therefore, the simulation results of the above method often cannot accurately reflect the dynamic response of the actual vehicle in a crosswind environment, resulting in lateral deviations that are inconsistent with the actual test or even directly driving out of the calculation domain, causing the simulation to terminate. There are defects such as a low degree of restoration of the actual situation, unreliable test results, and easy calculation failure. Summary of the invention

[0006] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes a closed-loop automobile crosswind stability dynamic evaluation system based on an MPC control algorithm, which has a high degree of restoration of the actual situation and improves the reliability of the test results and the success rate of the calculation.

[0007] In a first aspect, the present application provides a closed-loop automobile crosswind stability dynamic assessment system based on an MPC control algorithm, the system comprising: an interconnected CFD model unit, an MPC active control unit, and an MBD model unit;

[0008] The CFD model unit is used to obtain the aerodynamic six-component force parameters of the test vehicle according to the first motion state parameters of the test vehicle at the first moment and the side wind environment parameters;

[0009] The MPC active control unit is used to predict the state quantity and control quantity of the test vehicle at multiple time steps based on the MPC algorithm and with the goal of minimizing the control cost according to the first motion state parameter, so as to determine the front wheel steering angle parameter corresponding to the optimal control quantity of the test vehicle;

[0010] The MBD model unit is used to update the dynamic parameters of the test vehicle according to the six aerodynamic force parameters and the front wheel steering angle parameters, obtain the dynamic response parameters of the test vehicle at the second moment, and evaluate the crosswind stability of the test vehicle according to the dynamic response parameters.

[0011] According to one embodiment of the present application, the MPC active control unit includes: a vehicle body posture module, a vehicle body structure module, an error calculation module, an MPC algorithm module and a path planning module that are interconnected;

[0012] The vehicle body posture module is used to obtain the first motion state parameter of the test vehicle and the vehicle body posture parameter at the first moment;

[0013] The body structure module is used to obtain body structure parameters of the test vehicle;

[0014] The path planning module is used to obtain the planned path of the test vehicle;

[0015] The error calculation module is used to obtain a controller input according to the first motion state parameter, the body posture parameter at the first moment, the body structure parameter and the planned path;

[0016] The MPC algorithm module is used to iteratively optimize the control amount of the test vehicle according to the controller input amount with the goal of minimizing the control cost, so as to obtain the front wheel steering angle parameter at the second moment.

[0017] According to one embodiment of the present application, the MPC algorithm module is further used to:

[0018] Obtaining a state space expression of a state equation of the front wheel turning angle of the test vehicle;

[0019] Determining a state prediction and an output prediction of the test vehicle based on the state space expression and the controller input;

[0020] Based on the time control condition and the input human control condition of the test vehicle, construct a cost function of the control cost and a solution condition of the cost function;

[0021] According to the state prediction and the output prediction, the cost function is converted to obtain an equivalent cost function;

[0022] Based on the solution condition, solving the equivalent cost function to obtain the optimal state quantity;

[0023] The optimal control amount corresponding to the optimal state amount is used as the front wheel steering angle parameter at the second moment.

[0024] According to one embodiment of the present application, the cost function is:

[0025]

[0026] Where t-1 is the last sampling time, N p 、N c Respectively represent the prediction step size and control step size, y p (k+i|k), y ref (k+i|k) represents the predicted and reference values ​​of the control output, respectively, where i=1, 2, ...N p , u(k+i), Δu(k+i) represent the control input and control input increment at time k+i, respectively, where i=0,1,…,N c -1, Q and R are the system output, control increment and control weight system matrices respectively.

[0027] According to one embodiment of the present application, the solution condition is:

[0028] u min ≤u(k+i)≤u max , i=0,1,…N c -1

[0029] Δu min ≤Δu(k+i)≤Δu max , i=0,1,…N c -1

[0030] In the formula, u min 、u max , Δu min , Δu max They respectively represent the end values ​​of the range of acceleration and the end values ​​of the range of acceleration increment.

[0031] According to one embodiment of the present application, the MBD model unit includes a control quantity module, a motion state module and a response module connected to each other;

[0032] The control amount module is used to update the control amount of the test vehicle according to the front wheel angle parameter, and to update the aerodynamic load parameter of the test vehicle according to the aerodynamic six-component force parameter;

[0033] The motion state module is used to solve the DAE equation according to the updated aerodynamic load parameter and the updated control amount, update the first motion state parameter of the test vehicle to the second motion state parameter, and obtain the vehicle body posture parameter at the second moment;

[0034] The response module is used to generate a dynamic response parameter at the second moment according to the second motion state parameter and the vehicle body posture parameter at the second moment.

[0035] According to one embodiment of the present application, the CFD model unit includes an update module and an output module connected to each other;

[0036] The updating module is used to update the dynamic mesh model of the test vehicle based on the first motion state parameter and the side wind environment parameter, and calculate the aerodynamic load parameter of the test vehicle based on the updated dynamic mesh model;

[0037] The output module is used to output the aerodynamic load parameters as the aerodynamic six-component force parameters.

[0038] According to an embodiment of the present application, the update module includes: a boundary condition submodule and a moving grid submodule;

[0039] The boundary condition submodule is used to update the boundary condition of the test vehicle according to the first motion state parameter of the test vehicle and the side wind environment;

[0040] The dynamic mesh submodule is used to update the dynamic mesh model of the test vehicle based on the boundary conditions and the first motion state parameters, and calculate the aerodynamic load parameters through the dynamic mesh model.

[0041] According to an embodiment of the present application, the updating module is further used to update the first motion state parameter of the test vehicle according to the dynamic response parameter.

[0042] According to one embodiment of the present application, the response module can also perform response characteristic analysis, stability evaluation and dynamic stability evaluation based on the dynamic response parameters to obtain an evaluation result of the crosswind stability of the test vehicle.

[0043] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application.

[0044] The closed-loop automobile crosswind stability dynamic evaluation system based on the MPC control algorithm provided by the present invention has the following beneficial effects compared with the prior art:

[0045] (1) Based on the bidirectional dynamic coupling of vehicle aerodynamics and vehicle multibody dynamics (CFD-MBD), a CFD model and an MBD model for coupling calculation were established respectively to truly reflect the movement of the test vehicle in a crosswind environment, improve the degree of simulation restoration, and enhance the credibility of the test results. By considering the influence of lateral active control behavior in actual working conditions, the feedback effect of actual driving on crosswind interference is simulated, and the MPC algorithm is used in the dynamic coupling method to perform lateral control. According to the motion state of the current time step, the state parameters for a period of time in the future are predicted and the control output for a period of time in the future is calculated. The optimal control amount is selected as the front wheel turning angle parameter of the first control step, simulating the active control behavior of the driver or the role of the body electronic assistance system, forming a closed-loop evaluation, with a high degree of restoration of the actual situation, improving the reliability of the test results and the success rate of calculation, and being able to conduct low-cost and high-efficiency crosswind stability simulation tests to improve vehicle safety performance.

[0046] (2) The MPC algorithm is used for lateral control, so that the significant lateral deviation commonly seen in numerical simulations is effectively suppressed. It can also provide a closed-loop evaluation of the active correction behavior of the driver or electronic assistance system in a simulated crosswind environment, ensuring that the model does not undergo nearly uncontrolled lateral movement during solution, truly restoring the actual driving conditions, significantly improving the accuracy of numerical simulations in restoring actual driving conditions, and effectively avoiding simulation errors caused by large deviations. Compared with existing processing methods, it is more consistent with the actual situation. In the absence of crosswind stability test conditions, it truly reproduces the driving state of the car in a complex wind environment, quickly and effectively guides the forward development of the vehicle's aerodynamics and chassis, and provides a reference for the active control of intelligent driving cars in a crosswind environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0048] Figure 1 It is a structural schematic diagram of a closed-loop automobile sidewind stability dynamic evaluation system based on an MPC control algorithm provided in an embodiment of the present application;

[0049] Figure 2It is a schematic flow chart of a method executed by a closed-loop automobile sidewind stability dynamic evaluation system based on an MPC control algorithm provided in an embodiment of the present application;

[0050] Figure 3 It is a structural schematic diagram of a CFD model unit provided in an embodiment of the present application;

[0051] Figure 4 It is a schematic diagram of the structure of the MPC active control unit provided in an embodiment of the present application;

[0052] Figure 5 It is a structural diagram of the MBD model unit provided in the embodiment of the present application. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0054] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0055] In the following, in conjunction with the accompanying drawings, a closed-loop automobile sidewind stability dynamic assessment system based on an MPC control algorithm provided in an embodiment of the present application is described in detail through specific embodiments and their application scenarios.

[0056] It should be noted that the system is built based on computational fluid dynamics (CFD), model predictive control (MPC) and model-based design (MBD).

[0057] like Figure 1 As shown, the closed-loop automobile sidewind stability dynamic evaluation system based on the MPC control algorithm includes: a CFD model unit 110, an MPC active control unit 120 and an MBD model unit 130 connected to each other;

[0058] The CFD model unit 110 is used to obtain the aerodynamic six-component force parameters of the test vehicle according to the first motion state parameters of the test vehicle at the first moment and the side wind environment parameters;

[0059] The MPC active control unit 120 is used to predict the state quantity and control quantity of the test vehicle at multiple time steps based on the MPC algorithm and with the goal of minimizing the control cost according to the first motion state parameter, so as to determine the front wheel steering angle parameter corresponding to the optimal control quantity of the test vehicle;

[0060] The MBD model unit 130 is used to update the dynamic parameters of the test vehicle according to the six aerodynamic force parameters and the front wheel steering angle parameters, obtain the dynamic response parameters of the test vehicle at the second moment, and evaluate the crosswind stability of the test vehicle according to the dynamic response parameters.

[0061] The first moment may be the current moment, the second moment may be the predicted future moment, and the dynamic response parameters at the second moment include the second motion state parameters and the vehicle body posture parameters at the second moment.

[0062] The state variable is a variable that characterizes the state of the test vehicle, and the control variable is an input variable that affects the state of the test vehicle. The state variable can be used as a feedback input to adjust the control variable according to the current state.

[0063] The first motion state parameter and the second motion state parameter may include parameters such as position, speed, acceleration, heading angle, etc. of the test vehicle, which are used to describe the dynamic characteristics of the test vehicle at the first moment and the second moment, respectively.

[0064] The side wind environmental parameters may include the wind speed, wind direction, turbulence and vertical component of the side wind. The side wind not only affects the stability and handling of the vehicle, but may also affect safety.

[0065] The six-component aerodynamic parameters refer to the six main forces and moments that a car is subject to when it moves in an airflow in fluid dynamics. They can include aerodynamic force, lift, lateral force, rolling moment, pitch moment and yaw moment.

[0066] In actual implementation, Figure 2 As shown, the CFD model unit 110 solves the Navier-Stokes (NS) equations through the first motion state parameters and the side wind environment parameters to obtain the six-component aerodynamic force parameters of the test vehicle.

[0067] It should be noted that the study of crosswind stability of automobiles involves the intersection of multiple disciplines, such as automobile aerodynamics and automobile multibody dynamics. In order to comprehensively evaluate the crosswind stability of a vehicle, it is necessary to consider the mutual influence of factors such as aerodynamic disturbances, vehicle driving trajectory, and changes in body posture, and simulate the dynamic relationship between the six-component aerodynamic force parameters and body posture.

[0068] In related technologies, numerical simulation is used to more realistically obtain the flow field characteristics and motion state of the vehicle under crosswind conditions, which is used to test the stability of the car in a crosswind environment. In a strong crosswind environment, the simulated vehicle usually has a significant lateral deviation or even drives out of the calculation area because the lateral displacement obtained during the solution process is too large and lasts too long. This situation is quite different from the actual situation, which in turn affects the accuracy of the test results. After analysis, the reason is that the existing evaluation method of automobile crosswind stability ignores one point: the driver's active correction of direction and the role of auxiliary driving systems such as the body electronic stability system.

[0069] The existing two-way dynamic coupling method simulates the movement of the vehicle without corrective behavior in a crosswind environment by coupling vehicle aerodynamics and multi-body dynamics. However, this method ignores the driver's active control of the direction when the vehicle deviates from the path, and has certain limitations, which reduces the restoration accuracy of the evaluation and affects the credibility of the final evaluation results.

[0070] According to the closed-loop automobile crosswind stability dynamic evaluation system based on the MPC control algorithm provided in the embodiment of the present application, based on the bidirectional dynamic coupling of automobile aerodynamics and automobile multi-body dynamics (CFD-MBD), a CFD model and an MBD model for coupling calculation are respectively established to truly reflect the movement of the test vehicle in a crosswind environment, improve the degree of simulation restoration, and improve the credibility of the test results. By considering the influence of lateral active control behavior in actual working conditions, the feedback effect of actual driving on crosswind interference is simulated, and the MPC algorithm is used in the dynamic coupling method to perform lateral control. According to the motion state of the current time step, the state parameters for a period of time in the future are predicted and the control output for a period of time in the future is calculated. The optimal control amount is selected as the front wheel turning angle parameter of the first control step, and the driver's active control behavior or the role of the body electronic assistance system is simulated to form a closed-loop evaluation. The degree of restoration of the actual situation is high, the reliability of the test results and the success rate of the calculation are improved, and low-cost and high-efficiency crosswind stability simulation tests can be performed to improve vehicle safety performance.

[0071] The embodiments of the present application are closer to the actual situation. In the absence of crosswind stability test conditions, it can realistically reproduce the driving state of the vehicle in a complex wind environment, effectively prevent the MBD model from having a nearly uncontrolled lateral deviation during the solution process, truly restore the actual driving conditions, improve the degree of simulation restoration, and thereby improve the accuracy and reliability of the evaluation results.

[0072] In some embodiments, Figure 3 As shown, the CFD model unit 110 includes an update module 111 and an output module 112 connected to each other;

[0073] The updating module is used to update the dynamic mesh model of the test vehicle based on the first motion state parameter and the side wind environment parameter, and calculate the aerodynamic load parameter of the test vehicle based on the updated dynamic mesh model;

[0074] The output module is used to output the aerodynamic load parameters as the aerodynamic six-component force parameters.

[0075] Among them, the dynamic mesh model is the mesh used to simulate the interaction between fluid flow and object motion in computational fluid dynamics (CFD);

[0076] In actual execution, the update module obtains the dynamic response parameters of the first moment output by the MBD model unit 130 in real time, and obtains the first motion state parameters therefrom. In the update module, the geometric shape and distribution of the dynamic grid are adjusted according to the current position and posture of the test vehicle represented by the first motion state parameters to reflect the shape and movement of the test vehicle in a specific state. At the same time, according to the side wind environmental parameters, the influence of the side wind on the airflow field is considered, and the relative position or density of the grid nodes is adjusted to better adapt to the changes in external wind speed and wind direction.

[0077] Aerodynamic load parameters include lift, drag, lateral force, and moment on each axle of the test vehicle.

[0078] After the dynamic mesh model is updated, the model is used for CFD analysis to solve the NS equations to calculate the aerodynamic load parameters, which are then organized into six-component aerodynamic force parameters and output. The six-component aerodynamic force parameters are helpful in evaluating the aerodynamic characteristics and stability of the test vehicle in motion.

[0079] In addition, the CFD model unit also includes a determination module; the determination module is used to determine whether the simulation process of calculating the six-component aerodynamic force parameters is completed.

[0080] In some embodiments, the update module 111 includes: a boundary condition submodule and a moving grid submodule;

[0081] The boundary condition submodule is used to update the boundary condition of the test vehicle according to the first motion state parameter of the test vehicle and the side wind environment;

[0082] The dynamic mesh submodule is used to update the dynamic mesh model of the test vehicle based on the boundary conditions and the first motion state parameters, and calculate the aerodynamic load parameters through the dynamic mesh model.

[0083] It can be understood that the boundary conditions are used to define the relationship between the wind speed of the crosswind and time, and the updating of the boundary conditions is to ensure that the CFD simulation can accurately reflect the aerodynamic characteristics of the car in a specific state of motion.

[0084] The boundary conditions may include the front boundary, rear boundary, upper and lower boundaries, and side boundaries of the test vehicle.

[0085] In actual execution, according to the first motion state parameters and the lateral wind environment, the inflow velocity and wind speed are defined at the front boundary, and the actual inflow velocity is calculated. The rear boundary is set as a no-slip boundary to ensure fluid outflow. The upper and lower boundaries are set to symmetrical or no-slip boundary conditions, and the side boundaries are adjusted according to the wind speed and wind direction.

[0086] According to the first motion state parameter, the geometry of the grid is updated to the new position of the car, and according to the heading angle θ, the direction of the grid is adjusted to generate a new grid, and the grid is dynamically adjusted to adapt to the wind speed and direction considering the influence of the side wind.

[0087] In this embodiment, a mesh refinement method is used to increase the mesh density in key areas around the car, such as the front and sides, to improve calculation accuracy.

[0088] In some embodiments, the updating module is further used to update the first motion state parameter of the test vehicle according to the second motion state parameter.

[0089] The second motion state parameter is used to describe the motion state and behavior of the test vehicle after adjusting the front wheel steering angle parameter.

[0090] In some embodiments, Figure 4 As shown, the MPC active control unit 120 includes: a vehicle body posture module 121, a vehicle body structure module 122, a path planning module 123, an error calculation module 124 and an MPC algorithm module 125 which are interconnected;

[0091] The vehicle body posture module is used to obtain the first motion state parameter of the test vehicle and the vehicle body posture parameter at the first moment;

[0092] The body structure module is used to obtain body structure parameters of the test vehicle;

[0093] The path planning module is used to obtain the planned path of the test vehicle;

[0094] The error calculation module is used to obtain a controller input according to the first motion state parameter, the body posture parameter at the first moment, the body structure parameter and the planned path;

[0095] The MPC algorithm module is used to iteratively optimize the control amount of the test vehicle according to the controller input amount with the goal of minimizing the control cost, so as to obtain the front wheel steering angle parameter at the second moment.

[0096] It can be understood that the optimal control quantity may correspond to the optimal state quantity, which is obtained after evaluating multiple state quantities.

[0097] In actual execution, the vehicle body posture module 121 obtains in real time the dynamic response parameters at the first moment output by the MBD model unit 130 , where the motion response parameters at the first moment include the first motion state parameters and the vehicle body posture parameters.

[0098] The error calculation module integrates the first motion state parameter, the vehicle body posture parameter at the first moment, the vehicle body structure parameter and the planned path into a controller input and inputs the result into the MPC algorithm module.

[0099] The MPC algorithm module calculates the control quantity of multiple time steps and the state quantity corresponding to each control quantity according to the first motion state parameter in the controller input, the body posture parameter at the first moment, the body structure parameter and the planned path, and realizes cyclic optimization of the control quantity until the control quantity corresponding to the optimal state quantity is obtained, and the control quantity is used as the optimal control quantity, which is the front wheel steering angle parameter at the second moment.

[0100] In some embodiments, the MPC algorithm module 125 is further configured to:

[0101] Obtaining a state space expression of a state equation of the front wheel turning angle of the test vehicle;

[0102] Determining a state prediction and an output prediction of the test vehicle based on the state space expression and the controller input;

[0103] Based on the time control condition and the input human control condition of the test vehicle, construct a cost function of the control cost and a solution condition of the cost function;

[0104] According to the state prediction and the output prediction, the cost function is converted to obtain an equivalent cost function;

[0105] Based on the solution condition, solving the equivalent cost function to obtain the optimal state quantity;

[0106] The optimal control amount corresponding to the optimal state amount is used as the front wheel steering angle parameter at the second moment.

[0107] In actual implementation, the state equation of the front wheel steering angle is constructed through the MPC calculation module to control the error change of the front wheel steering angle. The state equation of the front wheel steering angle is expressed by the vehicle dynamics equation in the natural coordinate system as follows:

[0108]

[0109] In the formula, e d , Respectively represent the lateral error and heading error between the planned path and the actual path in the natural coordinate system, represents the first derivative of the lateral error, represents the first derivative of the lateral heading error, represents the second-order derivative of the lateral error, represents the second-order derivative of the lateral heading error, m represents the total mass of the vehicle, u represents the longitudinal speed at the center of mass, δ represents the wheel angle, I represents the moment of inertia, a and b represent the front and rear wheelbases, and K1 and K2 represent the cornering stiffness of the front and rear tires.

[0110] Among them, ed, and The calculation method is as follows:

[0111]

[0112] Where YO is the actual lateral position, YT is the target lateral position, is the yaw angle, XO is the actual longitudinal position, XT is the target longitudinal position; β is the sideslip angle of the center of mass, VX is the longitudinal vehicle speed; is the yaw rate, is the road curvature.

[0113] Furthermore, the forward Euler method is used to discretize the above vehicle dynamics equation to obtain the longitudinal discrete state equation:

[0114] x(k+1)=A k x(k)+B k u(k)

[0115] In the formula, x(k+1) and x(k) represent the state quantities at the current moment and the next moment, u(k) represents the control quantity at the current moment, that is, the front wheel steering angle parameter, and A k =E+A,B k =T s B, E, A, B are the unit matrix, the coefficient matrix before the state quantity and control quantity of the dynamic equation, respectively. s is the sampling period.

[0116] For the longitudinal discrete state equation, the state space expression of the state equation for the front wheel steering angle is constructed as follows:

[0117]

[0118] Where ξ(k|t)=[x(k)u(k-1)] T is the new state vector, and I m They represent the one-dimensional identity matrix and the two-dimensional identity matrix respectively, and m=1 and n=2.

[0119] Furthermore, through the rolling optimization of the MPC algorithm, the predicted state quantity and output quantity prediction can be expressed as:

[0120]

[0121] …

[0122]

[0123] …

[0124]

[0125] Furthermore, the state quantity prediction and output quantity prediction can be expressed in matrix form as:

[0126] Y=Ψξ(k)+ΘΔU

[0127] In the formula,

[0128] ΔU=[Δu(k) Δu(k+1)…Δu(k+N c )] T .

[0129] In some embodiments, the cost function is:

[0130]

[0131] Where t-1 is the last sampling time, N p 、N c Respectively represent the prediction step size and control step size, y p(k+i|k), y ref (k+i|k) represents the predicted and reference values ​​of the control output, respectively, where i=1, 2, ...N p , u(k+i), Δu(k+i) represent the front wheel steering angle parameter and the increment of the front wheel steering angle parameter at time k+i, respectively, where i = 0, 1, ..., N c -1, Q and R are the system output, control increment and control weight system matrices respectively.

[0132] In this embodiment, the cost function determined by control accuracy, avoiding excessive acceleration and excessive impact degree aims at minimizing the control cost, which can save the control resource consumption of the test vehicle.

[0133] In some embodiments, the solution condition is:

[0134] u min ≤u(k+i)≤u max , i=0,1,…N c -1

[0135] Δu min ≤Δu(k+i)≤Δu max , i=0,1,…N c -1

[0136] In the formula, u min 、u max , Δu min , Δu max They respectively represent the end values ​​of the range of acceleration and the end values ​​of the range of acceleration increment.

[0137] In this embodiment, the acceleration of the test vehicle and its rate of change are used as solution conditions for the constraint cost function, which can not only improve the driving safety of the vehicle, but also improve the overall performance of the control strategy.

[0138] Furthermore, the state quantity prediction and output quantity prediction are expressed in matrix form, and the cost function is converted. The equivalent cost function can be expressed as:

[0139]

[0140] In the formula, H = (Θ T Q Q Θ+R R ),

[0141] Furthermore, based on the equivalent cost function and solution conditions, the MPC solution problem is transformed into a standard quadratic programming problem:

[0142]

[0143] stΔU min ≤△U≤△U max

[0144] U min ≤A k ΔU+U t ≤U max

[0145] In the formula, The number of rows is N c The column vector of u(k-1) is the actual control value at the previous moment, U min , U max , ΔU min , ΔU max are the minimum, maximum and increment sets of the control quantity in the control time domain,

[0146] In this embodiment, the MPC algorithm is used to predict the future behavior of the test vehicle to optimize the control input to achieve the goal of minimizing the control cost, providing a basis for the dynamic evaluation of the crosswind stability of the test vehicle. The MPC algorithm runs synchronously with the two-way data exchange and provides feedback each time the state quantity is solved, thereby realizing a closed-loop dynamic evaluation of the vehicle's crosswind stability.

[0147] In some embodiments, Figure 5 As shown, the MBD model unit 130 includes a control quantity module 131, a motion state module 132 and a response module 133 which are interconnected;

[0148] The control amount module is used to update the control amount of the test vehicle according to the front wheel angle parameter, and to update the aerodynamic load parameter of the test vehicle according to the aerodynamic six-component force parameter;

[0149] The motion state module is used to solve the DAE equation according to the updated aerodynamic load parameter and the updated control amount, update the first motion state parameter of the test vehicle to the second motion state parameter, and obtain the vehicle body posture parameter at the second moment;

[0150] The response module is used to generate a dynamic response parameter at the second moment according to the second motion state parameter and the vehicle body posture parameter at the second moment.

[0151] Among them, aerodynamic load parameters are parameters related to the force and pressure exerted by the airflow acting on the surface of the object.

[0152] In actual execution, the control quantity module adjusts the test vehicle according to the front wheel steering angle parameters at the second moment to update the control quantity of the test vehicle, and adjusts the parameters of the aerodynamic model of the test vehicle according to the aerodynamic six-component force parameters to update the aerodynamic load parameters of the test vehicle. The motion state module constructs the vehicle motion equation according to the updated aerodynamic load parameters. The vehicle motion equation is used to provide the speed, acceleration and force conditions of the test vehicle at a certain moment. The motion state corresponding to the vehicle motion equation is updated by the updated control quantity to obtain a differential algebraic equation (DAE). By solving the equation, the second motion state parameters after the first motion state parameters are updated, as well as the body posture parameters at the second moment are obtained. The response module outputs the combination of the second motion state parameters and the body posture parameters at the second moment as the dynamic response parameters at the second moment to the update module 111 and the body posture module 121 to perform iteration for the next time step.

[0153] In some embodiments, the response module may also perform response characteristic analysis, stability evaluation, and dynamic stability evaluation based on the dynamic response parameters to obtain an evaluation result of the crosswind stability of the test vehicle.

[0154] In this embodiment, the MPC algorithm is used for lateral control, so that the significant lateral deviation commonly seen in numerical simulations is effectively suppressed, and the closed-loop evaluation of the active deviation correction behavior of the driver or the electronic assistance system in the simulated crosswind environment can be performed visually, ensuring that the model does not undergo nearly uncontrolled lateral movement during solution, and truly restores the actual driving conditions, significantly improving the accuracy of numerical simulation in restoring actual driving conditions, and effectively avoiding simulation errors caused by large deviations. Compared with existing processing methods, it is more consistent with the actual situation, and in the absence of crosswind stability test conditions, it truly reproduces the driving state of the car in a complex wind environment, quickly and effectively guides the forward development of the aerodynamics and chassis of the whole vehicle, and at the same time provides a reference for the active control of intelligent driving cars in crosswind environments.

[0155] The embodiments of the present application can solve the problem that the crosswind stability simulation test of an automobile in the prior art has a low degree of restoration of the actual situation, resulting in unreliable test results and prone to calculation failures, and provide a reference for the lateral control of an intelligent driving car in a crosswind environment, so as to truly restore the movement of the vehicle in a crosswind environment and reflect the impact of the active control behavior of the vehicle on the crosswind stability when a lateral deviation occurs.

[0156] In the description of this application, "first feature" or "second feature" may include one or more of the features.

[0157] In the description of the present application, “plurality” means two or more.

[0158] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

[0159] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0160] Although the embodiments of the present application have been shown and described, those skilled in the art will appreciate that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present application, and that the scope of the present application is defined by the claims and their equivalents.

Claims

1. A closed-loop vehicle sidewind stability dynamic assessment system based on MPC control algorithm, characterized in that: It includes a CFD model unit, an MPC active control unit and an MBD model unit connected to each other; The CFD model unit is used to obtain the aerodynamic six-component force parameters of the test vehicle according to the first motion state parameters of the test vehicle at the first moment and the side wind environment parameters; The MPC active control unit is used to predict the state quantity and control quantity of the test vehicle at multiple time steps based on the MPC algorithm and with the goal of minimizing the control cost according to the first motion state parameter, so as to determine the front wheel steering angle parameter corresponding to the optimal control quantity of the test vehicle; The MBD model unit is used to update the dynamic parameters of the test vehicle according to the aerodynamic six-component force parameters and the front wheel steering angle parameters, obtain the dynamic response parameters of the test vehicle at the second moment, and evaluate the crosswind stability of the test vehicle according to the dynamic response parameters.

2. The closed-loop automobile sidewind stability dynamic evaluation system based on the MPC control algorithm according to claim 1 is characterized in that: The MPC active control unit includes: a vehicle body posture module, a vehicle body structure module, an error calculation module, an MPC algorithm module and a path planning module which are interconnected; The vehicle body posture module is used to obtain the first motion state parameter of the test vehicle and the vehicle body posture parameter at the first moment; The body structure module is used to obtain body structure parameters of the test vehicle; The path planning module is used to obtain the planned path of the test vehicle; The error calculation module is used to obtain a controller input according to the first motion state parameter, the body posture parameter at the first moment, the body structure parameter and the planned path; The MPC algorithm module is used to iteratively optimize the control amount of the test vehicle according to the controller input amount with the goal of minimizing the control cost, so as to obtain the front wheel steering angle parameter at the second moment.

3. The closed-loop automobile sidewind stability dynamic evaluation system based on the MPC control algorithm according to claim 2 is characterized in that: The MPC algorithm module is further used to: Obtaining a state space expression of a state equation of the front wheel turning angle of the test vehicle; Determining a state prediction and an output prediction of the test vehicle based on the state space expression and the controller input; Based on the time control condition and the input human control condition of the test vehicle, construct a cost function of the control cost and a solution condition of the cost function; According to the state prediction and the output prediction, the cost function is converted to obtain an equivalent cost function; Based on the solution condition, solving the equivalent cost function to obtain the optimal state quantity; The optimal control amount corresponding to the optimal state amount is used as the front wheel steering angle parameter at the second moment.

4. The closed-loop automobile sidewind stability dynamic evaluation system based on the MPC control algorithm according to claim 3 is characterized in that: The cost function is: Where t-1 is the last sampling time, N p 、N c Respectively represent the prediction step size and control step size, y p (k+i|k), y ref (k+i|k) represents the predicted and reference values ​​of the control output, respectively, where i=1, 2, ...N p , u(k+i), Δu(k+i) represent the control input and control input increment at time k+i, respectively, where i=0, 1, ..., N c -1, Q and R are the system output, control increment and control weight system matrices respectively.

5. The closed-loop automobile sidewind stability dynamic evaluation system based on the MPC control algorithm according to claim 3 is characterized in that: The solution conditions are: u min ≤u(k+i)≤u max ,i=0,1,···N c -1 Δu min ≤Δu(k+i)≤Δu max ,i=0,1,…N c -1 In the formula, u min 、u max , Δu min , Δu max They respectively represent the end values ​​of the range of acceleration and the end values ​​of the range of acceleration increment.

6. The closed-loop automobile sidewind stability dynamic evaluation system based on the MPC control algorithm according to claim 1 is characterized in that: The MBD model unit includes a control module, a motion state module and a response module which are interconnected; The control amount module is used to update the control amount of the test vehicle according to the front wheel angle parameter, and to update the aerodynamic load parameter of the test vehicle according to the aerodynamic six-component force parameter; The motion state module is used to solve the DAE equation according to the updated aerodynamic load parameter and the updated control amount, update the first motion state parameter of the test vehicle to the second motion state parameter, and obtain the vehicle body posture parameter at the second moment; The response module is used to generate a dynamic response parameter at the second moment according to the second motion state parameter and the vehicle body posture parameter at the second moment.

7. The closed-loop automobile sidewind stability dynamic evaluation system based on the MPC control algorithm according to claim 1 is characterized in that: The CFD model unit includes an update module and an output module connected to each other; The updating module is used to update the dynamic mesh model of the test vehicle based on the first motion state parameter and the side wind environment parameter, and calculate the aerodynamic load parameter of the test vehicle based on the updated dynamic mesh model; The output module is used to output the aerodynamic load parameters as the aerodynamic six-component force parameters.

8. The closed-loop automobile sidewind stability dynamic evaluation system based on the MPC control algorithm according to claim 7 is characterized in that: The updating module includes: a boundary condition submodule and a moving grid submodule; The boundary condition submodule is used to update the boundary condition of the test vehicle according to the first motion state parameter of the test vehicle and the side wind environment; The dynamic mesh submodule is used to update the dynamic mesh model of the test vehicle based on the boundary conditions and the first motion state parameters, and calculate the aerodynamic load parameters through the dynamic mesh model.

9. The closed-loop automobile sidewind stability dynamic evaluation system based on the MPC control algorithm according to claim 7 is characterized in that: The updating module is further used to update the first motion state parameter of the test vehicle according to the dynamic response parameter.

10. The closed-loop automobile sidewind stability dynamic evaluation system based on the MPC control algorithm according to claim 6 is characterized in that: The response module can also perform response characteristic analysis, stability evaluation and dynamic stability evaluation according to the dynamic response parameters to obtain an evaluation result of the crosswind stability of the test vehicle.

Images