Intelligent chassis domain control method and system considering random disturbance, vehicle and medium

Abstract

The application relates to a smart chassis domain control method, system, vehicle and medium considering random interference. The chassis control domain method comprises the following steps: calculating a working condition expected value of a vehicle according to a reference track. A state equation with random disturbance is constructed, an information exchange mode of four subsystems of the vehicle is set, an optimal output sequence of a subsystem perturbation vector is solved through quadratic programming, an actual control amount is obtained, and finally, front and rear wheel steering angles, four-wheel driving torques and active suspension forces are obtained by distributing the actual control amount. Through the dynamic modeling of a two-freedom vehicle under a compound working condition, the theoretical reference value of the smart chassis is solved, the influence of external random disturbance on the execution subsystem of the smart chassis is considered, the dynamic state equation of the smart chassis is constructed, the actual control amount of the smart chassis is solved based on the cooperative game idea, and the driving stability and safety of the smart chassis are greatly improved.

Description

Technical Field

[0001] This invention relates to the field of vehicle control technology, and in particular to an intelligent chassis domain control method, system, vehicle, and medium that takes into account random disturbances. Background Technology

[0002] Intelligent drive-by-wire chassis are prone to functional overlap and control objective conflicts due to the independent control of each drive-by-wire subsystem, leading to the challenge of multi-system, multi-objective collaborative control of intelligent drive-by-wire chassis under complex operating conditions and external disturbances. In existing technologies, such as the invention patent with publication number CN121375749A entitled "A Multi-System Collaborative Game Control Method for Chassis Oriented to Path Tracking and Stability," the DYC (Direct Yaw Torque Control) system, AFS (Active Front Steering) system, ARS (Active Rear Steering) system, and ASS (Active Suspension System) system are treated as participants in a cooperative game. A total cost function is constructed, and the ADMM algorithm is used to solve the total cost function in a distributed manner to obtain the optimal control inputs for the DYC, AFS, ARS, and ASS systems. For example, the invention patent with publication number CN120003464A, entitled "A Cooperative Control Method for Distributed Drive Vehicle Chassis Based on Variable Steering Characteristics," generates the AFS front wheel steering angle and DYC additional yaw moment through MPC control. It distributes wheel drive or braking torque based on minimizing tire utilization and the target additional yaw moment, which is beneficial for setting constraints and improving control accuracy. It also improves vehicle lateral stability by minimizing tire utilization. Another invention patent, with publication number CN116424353A, entitled "A Coordinated Control Strategy for a Drive-by-Wire Chassis System Based on Distributed Vehicles," uses an adaptive steering wheel angle threshold to determine the vehicle's driving condition. The control strategy is adjusted in real time based on the determination result, employing a game-theoretic approach to coordinate the control of both, ensuring the vehicle's safety, stability, and fuel efficiency in dangerous situations. For example, the invention patent with publication number CN120620949A and invention titled "An Electric Vehicle Chassis Cooperative Control Method Considering Lateral and Roll Stability" applies robust model predictive control with online optimization solution. Based on the additional yaw moment and anti-roll moment output by the chassis cooperative controller, it achieves the optimized distribution of additional yaw moment by solving the problem of minimizing the utilization rate of four-wheel tire adhesion. Furthermore, it adjusts the distribution of anti-roll moment of the front and rear active suspensions based on the difference between the actual yaw rate and the reference value.

[0003] Clearly, existing intelligent drive-by-wire chassis domain control methods pay little attention to the integrated control of DYC, AFS, ARS, and ASS systems, and do not fully consider the dynamic coupling effect between longitudinal and lateral forces of the tires under combined operating conditions, making the vehicle more prone to critical states. Furthermore, existing research neglects the influence of external random disturbances caused by environmental changes and road conditions during actual vehicle operation, severely limiting the effectiveness of intelligent drive-by-wire chassis domain control algorithms in practical engineering applications. Summary of the Invention

[0004] To address the technical problem that existing vehicle chassis control systems do not consider the effects of complex operating conditions and external random disturbances, resulting in poor vehicle chassis driving stability, this invention provides an intelligent chassis domain control method, system, vehicle, and medium that considers random disturbances.

[0005] Firstly, this invention proposes an intelligent drive-by-wire chassis dynamics domain control method considering random disturbances, which is used to control four subsystems of the chassis. It includes the following steps: Based on the reference trajectory, the expected value x of the vehicle's operating condition is calculated considering parameters under combined operating conditions. ref In an external random perturbation environment, the input vector u of subsystem i is... i (k) is designed as: the steady-state feedback law of offline design and the perturbation vector c of online optimization. i The sum of (k+n|k). Construct a state equation considering external random perturbations, and based on the state equation and the perturbation vector c. i (k+n|k) calculates the optimal output sequence of subsystem i. : .

[0006] , .

[0007] .

[0008] .

[0009] In the formula, f, H, Θ i (k) is an intermediate quantity, C i (k) is the output sequence of subsystem i at time k. Ω is a constant. i R i Let Ψ be the weight matrix of subsystem i. i Let Λ be the input matrix of subsystem i. i Let z be the state matrix of subsystem i. i (k) is the nominal state vector of subsystem i, x iref(k) is the desired state vector of subsystem i, C j / i (k) represents the output sequence of subsystem i at time k, which is affected by subsystem j at time k-1. Let W be the perturbation matrix of subsystem i. i (k) represents the future N of the i-th subsystem. p A random perturbation sequence of N prediction steps. p Let i, j, d, and e be the labels of different subsystems, representing the total number of prediction steps. The perturbation vectors in the optimal output sequence of each subsystem are assigned to ensure that the vehicle's operating condition approximates x. ref This enables control of the four subsystems.

[0010] Secondly, this invention also proposes an intelligent drive-by-wire chassis dynamics domain control system that considers random disturbances, which uses the intelligent drive-by-wire chassis dynamics domain control method considering random disturbances from the first aspect. This intelligent drive-by-wire chassis dynamics domain control system includes: a prediction module, a solution module, and an output allocation module.

[0011] The prediction module is used to calculate the expected value x of the vehicle's operating conditions based on the reference trajectory and vehicle dynamics model, taking into account parameters of the combined operating conditions. ref The solver module is used to solve for the input vector u of subsystem i in an environment of external random perturbation. i (k) is designed as: the steady-state feedback law of offline design and the perturbation vector c of online optimization. i The sum of (k+n|k). Construct a state equation considering external random perturbations, and based on the state equation and the perturbation vector c. i (k+n|k) calculates the optimal output sequence of subsystem i. : .

[0012] , .

[0013] .

[0014] .

[0015] In the formula, f, H, Θ i (k) is an intermediate quantity, C i (k) is the output sequence of subsystem i at time k. Ω is a constant. i R i Let Ψ be the weight matrix of subsystem i. i Let Λ be the input matrix of subsystem i. i Let z be the state matrix of subsystem i. i(k) is the nominal state vector of subsystem i, x iref (k) is the desired state vector of subsystem i, C j / i (k) represents the output sequence of subsystem i at time k, which is affected by subsystem j at time k-1. Let W be the perturbation matrix of subsystem i. i (k) represents the future N of the i-th subsystem. p A random perturbation sequence of N prediction steps. p Let i, j, d, and e be the total number of prediction steps, and let i, j, d, and e be the labels of different subsystems. The output allocation module is used to allocate the perturbation vectors in the optimal output sequence of each subsystem to make the vehicle condition approximate x. ref This enables control of the four subsystems.

[0016] Thirdly, the present invention also proposes a distributed vehicle comprising the intelligent drive-by-wire chassis dynamics domain control system considering random disturbances as described in the second aspect, for controlling the chassis system of the distributed vehicle.

[0017] Fourthly, the present invention also proposes a computer-readable storage medium storing a computer program / instructions. When the computer program / instructions are executed by a processor, they implement the steps of the intelligent drive-by-wire chassis dynamics domain control method considering random disturbances in the first aspect.

[0018] The beneficial effects of this invention are as follows: This invention solves the theoretical reference value of the intelligent drive-by-wire chassis by modeling the two-free vehicle dynamics under combined working conditions. It also considers the impact of external random disturbances on the execution subsystem of the intelligent drive-by-wire chassis, constructs the dynamic state equation of the intelligent drive-by-wire chassis that takes into account external random disturbances, and solves the actual control quantity of the intelligent drive-by-wire chassis based on distributed stochastic model predictive control and cooperative game theory, which greatly improves the driving stability and safety of the intelligent drive-by-wire chassis. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of a dynamic domain control method for an intelligent drive-by-wire chassis that takes random disturbances into account. Figure 2 This is a comparison chart of the yaw rate of the experiment; Figure 3This is a comparison chart of the lateral displacement deviations in the experiment; Figure 4 This is a comparison chart of the side tilt angles of the experiment. Detailed Implementation

[0021] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the specification of this invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "or / and" as used herein includes any and all combinations of one or more of the associated listed items.

[0023] Please refer to Figure 1 This embodiment provides an intelligent drive-by-wire chassis dynamics domain control method that considers random disturbances. It mainly calculates the vehicle's expected operating value x based on the reference trajectory using the upper-level system's pre-aiming model and a two-degree-of-freedom vehicle dynamics model under combined operating conditions. ref The middle-level system collects information such as front and rear wheel steering angles, yaw rate, and roll angle through onboard sensors and establishes a four-degree-of-freedom vehicle dynamics model. Then, it constructs state equations with random perturbations and sets up information exchange modes for four subsystems: AFS, ARS, DYC, and ASS. The optimal output sequence of perturbation vectors of the four subsystems is obtained through quadratic programming, thereby obtaining the actual control quantities, namely front and rear wheel steering angles, additional yaw moment, and roll moment. Finally, the actuators of the lower-level system obtain the front and rear wheel steering angles, four-wheel drive torque, and active suspension force from the four subsystems.

[0024] Specifically, the upper-level system needs to calculate the expected operating value x of the vehicle. ref Includes: desired lateral position Y ref Desired lateral velocity v yref Expected yaw angle Desired yaw rate γ ref Desired roll angle Desired roll rate This can be represented as: The aiming model expression in the upper-level system is: .

[0025] .

[0026] .

[0027] .

[0028] In the formula, e y This represents the lateral tracking error. X0 is the ordinate of the vehicle's current position. Y0 is the abscissa of the vehicle's current position. k Y represents the ordinate of the reference trajectory preview point. k The x-coordinate of the reference trajectory preview point. Y is the yaw angle. Y is the lateral displacement of the vehicle. Δt is the prediction time. , These are the first and second derivatives of Y0, respectively. L is the optimal radius of curvature for the reference trajectory. p δ is the preview distance for the reference trajectory. ref To determine the desired front wheel steering angle, v x For the longitudinal speed of the vehicle, For the front wheel lateral stiffness under combined working conditions, denoted as Rear wheel lateral stiffness under combined working conditions, L as the front and rear wheelbase, m as the vehicle mass, a as the horizontal distance from the front axle to the vehicle's center of gravity, and b as the horizontal distance from the rear axle to the vehicle's center of gravity.

[0029] The vehicle dynamics model in the upper-level system adopts a two-degree-of-freedom model and considers parameters under combined working conditions. Its expression is as follows: .

[0030] .

[0031] ; .

[0032] ; .

[0033] .

[0034] In the formula, v y Let F be the vehicle's lateral velocity, γ be the yaw rate, and F be the lateral velocity. yf F is the lateral force on the front wheel. yr For the lateral force of the rear wheel, I z Let α be the yaw moment of inertia of the vehicle. f The front wheel slip angle, α r The rear wheel slip angle, μ cf μ cr All are intermediate variables, K fK represents the front wheel lateral stiffness under pure lateral slip conditions. r μ is the rear wheel lateral stiffness under pure lateral slip conditions. f μ is the coefficient of adhesion for the front wheels. r P is the rear wheel adhesion coefficient, P is the front and rear brake power distribution coefficient, and a is the rear wheel adhesion coefficient. x δ is the longitudinal acceleration, g is the gravitational acceleration, and h is the height of the vehicle's center of gravity. f This is the front wheel steering angle. δ r β is the rear wheel steering angle. β is the sideslip angle, which is: β = v y / v x Let the vehicle's lateral velocity v y First derivative The first derivative of γ Substituting the values ​​into the vehicle dynamics model, the desired yaw rate γ is calculated. ref : .

[0035] In the formula, min(·) is the minimum function, and sign(·) is the sign function. K is the stability factor, an important parameter characterizing the steady-state response of a vehicle. In this embodiment, μ is the road adhesion coefficient. Finally, let the desired lateral velocity... Desired roll angle Desired roll rate Furthermore, the expected operating value x of the vehicle can be calculated using the pre-aiming model and the vehicle dynamics model. ref .

[0036] In the middle layer system, a vehicle dynamics model considering the coupling effects of roll and lateral dynamics is constructed, which can be a four-degree-of-freedom model: .

[0037] In the formula, M z This is for the additional yaw moment. Φ is the vehicle roll angle. (m) s h represents the sprung mass of the vehicle. s This is the horizontal distance from the vehicle's center of gravity to its roll center. x C represents the moment of inertia of the vehicle's sprung mass about the direction of travel. Φ K represents the total roll angle damping of the vehicle. Φ M represents the total roll stiffness of the vehicle. x This is the additional roll moment for the active suspension system.

[0038] Then, the state vector x is introduced: Construct discretized state equations considering external random disturbances. The four-degree-of-freedom vehicle dynamics model is transformed into: .

[0039] .

[0040] .

[0041] .

[0042] .

[0043] . .

[0044] In the formula, Let x be the first derivative of the state vector. A is the state matrix. B i Let u be the control matrix of the i-th subsystem. i Let be the input vector of the i-th subsystem. Let y be the output vector. Let C be the observation matrix. Let G be the perturbation matrix that determines the effect of random perturbations. i Let be the external random perturbation vector of the i-th subsystem. The state equation is then transformed into a discrete form using the forward Euler method: .

[0045] Where x(k+1) is the discrete state vector x at time k+1, A d Let A be the discrete state matrix, and A d =A+AT. B di The discretized control matrix B i And B di =B i T. u i (k) is the discrete input vector u at time k. i G d Let G be the discretized perturbation matrix, and G d =GT. w i (k) is the discrete external random disturbance vector w at time k. i y(k) is the discrete output vector y at time k. C d Let C be the discretized observation matrix. x(k) is the discretized state vector x at time k. T is the sampling time. One innovation of this invention is considering the influence of external random disturbances. These external random disturbances are represented by the external random disturbance term G. d w i (k) is mathematically described. Here, the external random perturbation vector w i (k) is the disturbance vector formed by external random disturbance. In the calculation, it is assumed that the external random disturbance vector is w. i (k) All components have a mean of zero and are independent of each other, and satisfy the following condition: .

[0046] .

[0047] In the formula, Pr{·} represents the probability of the event occurring. F i (·) represents the distribution function. η i For the distribution function variable. It is a finite set of support values. The finite support value for the nth prediction step is the upper bound of the customizable perturbation vector. It is a multicellular organism. For defining symbols. N p This represents the total number of prediction steps.

[0048] On the other hand, to address the information exchange issue among the various drive-by-wire subsystems of the intelligent drive-by-wire chassis, this embodiment sets up an information exchange mode between the four subsystems: at sampling time k, subsystem i shares the control strategy information of other subsystems at sampling time k-1. Specifically, at sampling time k, subsystem i shares the control strategy with subsystem j via the Controller Area Network (CAN) bus, which is controlled by subsystem i's control strategy U at sampling time k-1. i (k-1) constructs can be represented as U i / j (k). Therefore, when solving the optimization problem individually at time k, the shared information can be considered as a known sequence of fixed constants of the other subsystems. Using Pareto optimality theory, the overall system cost of the distributed system is minimized by simultaneously solving the optimization problems of the four subsystems.

[0049] After determining the state equation of the chassis and the communication strategy among the four subsystems, under an external random disturbance environment, the input vector u of the subsystems is... i (k) is designed as: offline design of the steady-state feedback law K of the i-subsystem. i x i (k+n|k) and the perturbation vector c of the online optimized i-subsystem i The sum of (k+n|k), i.e. When n≥N p At that time, the perturbation vector c i (k+n|k)=0. Subsequently, the optimal output sequence of each subsystem is calculated based on the state equations and the decomposed perturbation vectors. The specific solution process is as follows: The state vector x of subsystem i i For example, let's break it down into two parts: .in, In nominal condition, and . The part is uncertain, and Intermediate quantity Φ: Φ = Ad +B di K i K i This is the linear state feedback matrix for the offline-confirmed steady-state feedback law. When the prediction step n=0, the initial condition is z. i (k)=x i (k), e i (k)=0.

[0050] When the input vector u i (k) and state vector x i After decomposition, the resulting perturbation vector c i (k) and nominal state z i (k) Substitute the discretized state equations into the prediction equations to predict the dynamic state response of the intelligent drive-by-wire chassis. The prediction equations are: .

[0051] Among them, Z i (k) is the vehicle nominal state prediction sequence at sampling time k, and Γ i Let N be the desired state matrix of subsystem i. p This represents the total number of prediction steps. i (k) represents the nominal state of the vehicle at sampling time k. Λ i Let be the state matrix, and C i (k) is the perturbation vector sequence of subsystem i at sampling time k, and c i (k+n|k) is the perturbation vector of subsystem i. Ψ i Let be the input matrix of subsystem i, and B di B is the control matrix after discretization of subsystem i. i . The perturbation matrix determines the mode of action of random perturbations, and G di Let G be the perturbation matrix of subsystem i. d W i (k) represents the future N of the i-th subsystem. p A random perturbation sequence for each prediction step, and .

[0052] In addition, the following probability constraints are set for the state variables of each drive-by-wire subsystem of the intelligent drive-by-wire chassis: .

[0053] In the formula, Pr{·} represents the probability of the event occurring. max(k) represents the maximum value of the vehicle state variables. P c This is a probability. This probability constraint can be transformed into a nominal state. ,Right now: .

[0054] in, For actuator constraints. i_sat To constrain the amount of shrinkage.

[0055] Meanwhile, to address the external random disturbance problem faced by the intelligent drive-by-wire chassis, a cost function is constructed to solve for its control variables, i.e., to find the optimal output sequence of each subsystem. First, the cost function for each subsystem is constructed: AFS subsystem cost function J AFS The expression for (k) is: .

[0056] in, . , .

[0057] Similarly, the cost function J of the DYC subsystem DYC The expression for (k) is: .

[0058] in, . , .

[0059] The cost function J of the ASS subsystem ASS The expression for (k) is: .

[0060] in, . , .

[0061] The cost function J of the ARS subsystem ARS The expression for (k) is: .

[0062] in, . , .

[0063] In the formula, since the control method in this embodiment is for four chassis systems, i is taken as: i=1,2,3,4, which respectively refer to the four subsystems. The subscript i=1 represents the AFS subsystem, i=2 represents the DYC subsystem, i=3 represents the ASS subsystem, and i=4 represents the ARS subsystem. Let r1, r2, r3, r4, Q1, Q2, Q3, and Q4 be the expected state matrix of the corresponding subsystem. Let r1, r2, r3, r4, Q1, Q2, Q3, and Q4 be the positive semi-definite weight matrices of the corresponding subsystem. 1ref x 2ref x 3ref x 4ref These are the expectation matrices of the corresponding subsystems. Let be the desired yaw rate. This is the horizontal position weighting coefficient. This is the yaw angle weighting coefficient. This is the lateral velocity weighting coefficient. This is the yaw rate weighting coefficient. This is the roll angle weighting coefficient. This is the weighting coefficient for the roll rate.

[0064] Next, to solve the multi-subsystem, multi-objective cooperative control problem of intelligent drive-by-wire chassis, cooperative game theory is used to solve it. First, based on the global information sharing among the drive-by-wire subsystems of the intelligent drive-by-wire chassis, , , , The individual interests of the four subsystems, as game participants, are reformulated as a linear weighted cost function: .

[0065] In the formula, J sum (k) J sum_i (k) is the total cost function of subsystem i. λ1(k), λ2(k), λ3(k), and λ4(k) are the relative weights of the cost function of the corresponding subsystem at sampling time k. They can be dynamically adjusted according to the vehicle state, satisfying λ1(k) + λ2(k) + λ3(k) + λ4(k) = 1, and each relative weight is greater than 0.

[0066] Subsequently, to quickly find the optimal output sequence, a quadratic programming optimization method is used to calculate the optimal output sequence for each subsystem. Specifically: First, to efficiently find the optimal output sequence, based on the matrices and prediction states within the prediction equation, the problem of finding the optimal output sequence can be solved as a quadratic programming (QP) problem. This linearly weighted cost function can be rewritten in the form of a quadratic programming problem: .

[0067] In the formula, ||·|| 2 Let η be the Euclidean norm, and η be a constant. Weight matrix. Q i The weight matrix is ​​positive semi-definite. . λi These represent the relative weights of the cost function for the corresponding subsystem. Weight matrix. Weight matrix .

[0068] Next, we use quadratic programming optimization methods to solve the above quadratic programming problem: .

[0069] , .

[0070] .

[0071] .

[0072] This is the optimal output sequence for subsystem i. The optimal output sequence is used to make the vehicle operating condition approximate the desired operating condition value x. ref In the formula, f, H, and Θ i (k) is an intermediate quantity, C i (k) is the output sequence of subsystem i at time k. As a negligible intermediate quantity, it is Ω i R i Let Ψ be the weight matrix of subsystem i. i Let Λ be the input matrix of subsystem i. i Let z be the state matrix of subsystem i. i (k) is the nominal state vector of subsystem i, x iref (k) is the desired state vector of subsystem i, C j / i (k) represents the output sequence of subsystem i at time k, which is affected by subsystem j at time k-1. Similarly, i, j, d, and e are the labels of different subsystems, then Ψ d Let Ψ be the input matrix of the d subsystem. e Let C be the input matrix of the e-subsystem. d / i (k) represents the output sequence of subsystem i at time k, which is affected by subsystem d at time k-1. e / i (k) is the output sequence of subsystem i at time k, which is affected by subsystem e at time k-1. Let W be the perturbation matrix of subsystem i. i (k) represents the future N of the i-th subsystem. p A sequence of random perturbations for each prediction step.

[0073] After obtaining the optimal output sequence of each subsystem, i.e., the actual control input of the intelligent drive-by-wire chassis, the output is distributed through the actuators of the lower-level system. Specifically, for the ARS and AFS subsystems, the rear wheel steering angle δ rand front wheel steering angle δ f This will be performed by the steer-by-wire motor. For the DYC subsystem, since the four wheels of the distributed electric drive vehicle are independently controllable, the additional yaw torque M generated by the hub motors needs to be controlled. z The force is rationally distributed to each wheel. To ensure full utilization of the adhesion between the tire and the road surface under extreme conditions, the optimized distribution strategy aims to minimize tire utilization, i.e., minimize the ratio of tire force to maximum adhesion. This is based on the overall longitudinal force F. x and additional yaw moment M z The following equation can be established: .

[0074] In the formula, R w T represents the tire radius. s This represents the torque of the s-th hub motor. These correspond to the left front wheel, right front wheel, left rear wheel, and right rear wheel, respectively. f It is half the track width of the front wheels. d r It is half the distance of the rear wheel track.

[0075] To minimize the utilization rate of each tire, the cost function J tire Designed as follows: .

[0076] In the formula, F zs Let T be the vertical force of the s-th tire. s The constraints are: .

[0077] Where T min and T max These are the minimum and maximum output torques for each hub motor, respectively.

[0078] Finally, the optimal torque distribution problem can be expressed in the form of quadratic programming as follows: .

[0079] In the formula, the torque matrix Weight matrix The final calculation yielded the output torque of the four wheel hub motors.

[0080] For the ASS subsystem, since the distance 'a' from the center of gravity to the front axle is not equal to the distance 'b' from the center of gravity to the rear axle, an additional roll moment M is required. x The distribution principle requires maximizing roll resistance without affecting vehicle pitch motion. The corresponding additional active suspension force can be calculated as follows: .

[0081] In the formula, F l1 For the additional active suspension force on the left front wheel, F l2 This is the additional active suspension force for the left rear wheel. F r1 For the additional active suspension force on the right front wheel, F r2 This is the additional active suspension force for the right rear wheel.

[0082] To demonstrate the effectiveness of the intelligent drive-by-wire chassis dynamics domain control method considering random disturbances, an experiment was conducted. The experiment used a dry road surface with two lane change maneuvers, setting the road adhesion coefficient to 0.85 and the vehicle speed to 72 km / h. The experimental results are as follows: Figure 2 , Figure 3 , Figure 4 As shown. By Figure 2 It can be seen that the control method of the present invention reduces the root mean square error (the square root of the average of the squares of the deviations between the observed and expected values) of yaw rate by 86.77% compared to no control, and by 46.01% compared to the existing active front wheel steering angle control, demonstrating a significant tracking effect on the desired yaw rate. Figure 3 It can be seen that the control method of the present invention reduces the root mean square error of lateral displacement deviation by 96.06% compared to no control, and by 75.34% compared to the existing active front wheel steering angle control, greatly improving the vehicle's trajectory tracking accuracy. Figure 4 It can be seen that the root mean square error of the roll angle of the control method of the present invention is reduced by 32.97% compared with no control and by 54.48% compared with the active front wheel steering angle control in the prior art, effectively improving roll safety.

[0083] In another embodiment, an intelligent drive-by-wire chassis dynamics domain control system considering random disturbances is proposed, which uses the intelligent drive-by-wire chassis dynamics domain control method considering random disturbances described in the above embodiments. This intelligent drive-by-wire chassis dynamics domain control system includes: a prediction module, a solution module, and an output allocation module.

[0084] The prediction module is used to calculate the expected value x of the vehicle's operating conditions based on the reference trajectory and vehicle dynamics model, taking into account parameters of the combined operating conditions. ref The solver module is used to solve for the input vector u of subsystem i in an environment of external random perturbation. i (k) is designed as: the steady-state feedback law of offline design and the perturbation vector c of online optimization. i The sum of (k+n|k). Construct a state equation considering external random perturbations, and based on the state equation and the perturbation vector c. i (k+n|k) calculates the optimal output sequence of subsystem i. : .

[0085] , .

[0086] .

[0087] .

[0088] In the formula, f, H, Θ i (k) is an intermediate quantity, C i (k) is the output sequence of subsystem i at time k. Ω is a constant. i R i Let Ψ be the weight matrix of subsystem i. i Let Λ be the input matrix of subsystem i. i Let z be the state matrix of subsystem i. i (k) is the nominal state vector of subsystem i, x iref (k) is the desired state vector of subsystem i, C j / i (k) represents the output sequence of subsystem i at time k, which is affected by subsystem j at time k-1. Let W be the perturbation matrix of subsystem i. i (k) represents the future N of the i-th subsystem. p A random perturbation sequence of N prediction steps. p Let i, j, d, and e be the total number of prediction steps, and let i, j, d, and e be the labels of different subsystems. The output allocation module is used to allocate the perturbation vectors in the optimal output sequence of each subsystem to make the vehicle condition approximate x. ref This enables control of the four subsystems.

[0089] In another embodiment, a distributed vehicle is proposed, which includes the intelligent drive-by-wire chassis dynamics domain control system that considers random disturbances as described in the above embodiments, for controlling the chassis system of the distributed vehicle.

[0090] In another embodiment, a computer-readable storage medium is also proposed, which stores a computer program / instructions. When the computer program / instructions are executed by a processor, the steps of the intelligent drive-by-wire chassis dynamics domain control method considering random disturbances in the above embodiments are implemented. The computer-readable storage medium may include, but is not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above.

[0091] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0092] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A dynamic domain control method for an intelligent drive-by-wire chassis considering random disturbances, used to control four subsystems of the chassis, characterized in that, It includes: calculating the expected value x of the vehicle in the operating mode on the basis of the reference trajectory under consideration of the parameters of the composite operating mode ref ; In an external random perturbation environment, the input vector u of subsystem i... i (k) is designed as: the steady-state feedback law of offline design and the perturbation vector c of online optimization. i The sum of (k+n|k); Construct a state equation that considers external random perturbations, and based on the state equation and the perturbation vector c i (k+n|k) calculates the optimal output sequence of subsystem i. : ； ， ； ； ； In the formula, f, H, Θ i (k) is an intermediate quantity, C i (k) is the output sequence of subsystem i at time k. Ω is a constant. i R i Let Ψ be the weight matrix of subsystem i. i Let Λ be the input matrix of subsystem i. i Let z be the state matrix of subsystem i. i (k) is the nominal state vector of subsystem i, x iref (k) is the desired state vector of subsystem i, C j / i (k) represents the output sequence of subsystem i at time k, which is affected by subsystem j at time k-1. Let W be the perturbation matrix of subsystem i. i (k) represents the future N of the i-th subsystem. p A random perturbation sequence of N prediction steps. p The total number of prediction steps is represented by i, j, d, and e, which are the labels of different subsystems, respectively. The perturbation vectors in the optimal output sequence of each subsystem are allocated to make the vehicle operating condition approximate x. ref This enables control of the four subsystems.

2. The intelligent drive-by-wire chassis dynamics domain control method considering random disturbances according to claim 1, characterized in that, Calculate the expected value x under the operating conditions ref At that time, the reference trajectory is tracked through a preview model and a vehicle dynamics model; the preview model expression is: ； ； ； ； In the formula, e y For lateral tracking error, X0 is the ordinate of the vehicle's current position, Y0 is the abscissa of the vehicle's current position, and X... k The ordinate of the reference trajectory preview point is Y. k The x-coordinate of the reference trajectory preview point position. Let Y be the yaw angle, Y be the lateral displacement of the vehicle, and Δt be the prediction time. L is the optimal radius of curvature for the reference trajectory. p δ is the preview distance for the reference trajectory. ref To determine the desired front wheel steering angle, v x For the longitudinal speed of the vehicle, For the front wheel lateral stiffness under combined working conditions, denoted as Rear wheel lateral stiffness under combined working conditions, L as the front and rear wheelbase, m as the vehicle mass, a as the horizontal distance from the front axle to the vehicle's center of gravity, and b as the horizontal distance from the rear axle to the vehicle's center of gravity.

3. The intelligent drive-by-wire chassis dynamics domain control method considering random disturbances according to claim 2, characterized in that, The vehicle dynamics model considers parameters under combined operating conditions, and its expression is as follows: ； ； ； ； ； ； In the formula, v y Let F be the vehicle's lateral velocity, γ be the yaw rate, and F be the lateral velocity. yf F is the lateral force on the front wheel. yr For the lateral force of the rear wheel, I z Let α be the yaw moment of inertia of the vehicle. f The front wheel slip angle, α r The rear wheel slip angle, μ cf μ cr K is an intermediate variable. f K represents the front wheel lateral stiffness under pure lateral slip conditions. r μ is the rear wheel lateral stiffness under pure lateral slip conditions. f μ is the coefficient of adhesion for the front wheels. r P is the rear wheel adhesion coefficient, P is the front and rear brake power distribution coefficient, and a is the rear wheel adhesion coefficient. x ρ is the longitudinal acceleration, g is the gravitational acceleration, and h is the height of the center of mass.

4. The intelligent drive-by-wire chassis dynamics domain control method considering random disturbances according to claim 1, characterized in that, Expected value of operating conditions x ref Includes: desired lateral position Y ref Desired lateral velocity v yref Expected yaw angle Desired yaw rate γ ref Desired roll angle Desired roll rate ; in, ; , ; And / or, considering external random disturbances, the state equation is: ； x(k+1) is the state vector x at time k+1, A d Let B be the state matrix. di For the control matrix, G d Let w be the perturbation matrix. i y(k) is the external random perturbation vector at time k, y(k) is the output vector at time k, and C d This is the observation matrix.

5. The intelligent drive-by-wire chassis dynamics domain control method considering random disturbances according to claim 1, characterized in that, In the process of calculating the optimal output sequence of each subsystem, the information exchange mode of the four subsystems is set: at sampling time k, subsystem i shares the control strategy with subsystem j through the controller area network bus, which is the control strategy U of subsystem i at sampling time k-1. i The (k-1)-th subsystem is constructed, and its control strategy at sampling time k is represented as U. i / j (k).

6. The intelligent drive-by-wire chassis dynamics domain control method considering random disturbances according to claim 1, characterized in that, Calculate the optimal output sequence of subsystem i The methods include: Construct the optimal output sequence for solving the i-subsystem The cost function of the subsystem; The state vector x of subsystem i i Decomposition yields the nominal state z i (k); the perturbation vector c i (k+n|k) and nominal state z i (k) Substitute the state equations to establish the prediction equations for the dynamic state response of the intelligent drive-by-wire chassis: ； In the formula, Z i (k) is the vehicle nominal state prediction sequence at sampling time k, Λ i Let Ψ be the state matrix. i Given the input matrix, C i (k) is the perturbation vector sequence of subsystem i at sampling time k. The perturbation matrix W determines the manner in which random perturbations occur. i (k) is a random perturbation sequence; Then, based on the matrices in the prediction equation and the prediction state, the cost function is rewritten as a quadratic programming problem. The quadratic programming optimization method is used to solve the quadratic programming problem in order to obtain the optimal output sequence of each subsystem.

7. The intelligent drive-by-wire chassis dynamics domain control method considering random disturbances according to claim 1, characterized in that, The four subsystems are: ARS subsystem, AFS subsystem, DYC subsystem, and ASS subsystem; When assigning perturbation vectors to each subsystem, the perturbation vector c corresponding to the ARS subsystem and the AFS subsystem is... i (k+n|k) represent the rear wheel steering angle and the front wheel steering angle, respectively; The perturbation vector c corresponding to the DYC subsystem i (k+n|k) represents the additional yaw moment M. z ; and with minimizing tire utilization as the optimization objective, based on the vehicle's longitudinal force F x and additional yaw moment M z Calculate the output torque of the four hub motors; The perturbation vector c corresponding to the ASS subsystem i (k+n|k) represents the additional roll moment M. x ; and by adding a roll moment M x Calculate the additional active suspension force for all four tires.

8. A smart drive-by-wire chassis dynamics domain control system considering random disturbances, characterized in that, It uses the intelligent drive-by-wire chassis dynamics domain control method considering random disturbances as described in any one of claims 1 to 7; the control system includes: The prediction module is used to calculate the expected value x of the vehicle's operating condition based on the reference trajectory and vehicle dynamics model, taking into account parameters of combined operating conditions. ref ; The solver module is used to solve for the input vector u of subsystem i in an external random perturbation environment. i (k) is designed as: the steady-state feedback law of offline design and the perturbation vector c of online optimization. i The sum of (k+n|k); construct a state equation considering external random perturbations, and based on the state equation and the perturbation vector c i (k+n|k) calculates the optimal output sequence of subsystem i. : ； ， ； ； ； In the formula, f, H, Θ i (k) is an intermediate quantity, C i (k) is the output sequence of subsystem i at time k. Ω is a constant. i R i Let Ψ be the weight matrix of subsystem i. i Let Λ be the input matrix of subsystem i. i Let z be the state matrix of subsystem i. i (k) is the nominal state vector of subsystem i, x iref (k) is the desired state vector of subsystem i, C j / i (k) represents the output sequence of subsystem i at time k, which is affected by subsystem j at time k-1. Let W be the perturbation matrix of subsystem i. i (k) represents the future N of the i-th subsystem. p A random perturbation sequence of N prediction steps. p The total number of prediction steps is represented by i, j, d, and e, which are the labels of different subsystems, respectively. The output allocation module is used to allocate the perturbation vectors in the optimal output sequence of each subsystem so that the vehicle operating condition approximates x. ref This enables control of the four subsystems.

9. A distributed vehicle, characterized in that, It includes the intelligent drive-by-wire chassis dynamics domain control system as described in claim 8, for controlling the chassis system of a distributed vehicle.

10. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instruction is executed by the processor, it implements the steps of the intelligent drive-by-wire chassis dynamics domain control method considering random disturbances as described in any one of claims 1 to 7.

Images