Vehicle lateral control method and device, computer device and storage medium

Abstract

The application relates to the field of automatic driving of an automobile, and discloses a vehicle lateral control method and device, computer equipment and a storage medium, the method comprising the following steps: acquiring vehicle preset parameters, space prediction parameters and prediction custom parameters, and configuring a state prediction model according to the prediction custom parameters; processing the vehicle preset parameters and the space prediction parameters through the state prediction model to obtain a prediction state quantity; acquiring a path reference sequence, determining a position prediction error according to the path reference sequence and the prediction state quantity; acquiring a custom weighting matrix, and determining a control gain according to the custom weighting matrix and the prediction state quantity; determining a prediction control increment according to the position prediction error and the control gain; acquiring a current control quantity, and determining a target control quantity according to the prediction control increment and the current control quantity. The application can improve the accuracy and stability of vehicle lateral control.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving, and more particularly to a vehicle lateral control method, device, computer equipment, and storage medium. Background Technology

[0002] In existing technologies, vehicle lateral control methods mainly employ PID control (proportional, integral, and derivative control), linear quadratic programming control, and traditional MPC control (model predictive control).

[0003] Among these, PID control is a model-free control method, characterized by its simplicity, few parameters, and high practicality. However, it lacks effective feedforward compensation for both measurable and unmeasurable disturbances in the system, resulting in instability, poor control accuracy, and a lack of effective utilization of future and past states. Linear quadratic programming control is a globally optimal control method, requiring the solution of the control sequence over the entire prediction time domain, leading to a large computational burden. Traditional MPC control methods require accurate model parameters during modeling, which is challenging to obtain. Summary of the Invention

[0004] Therefore, it is necessary to provide a vehicle lateral control method, device, computer equipment, and storage medium to address the aforementioned technical problems, so as to improve the accuracy and stability of vehicle lateral control.

[0005] A vehicle lateral control method, comprising:

[0006] Obtain vehicle preset parameters, spatial prediction parameters, and prediction custom parameters, and configure a state prediction model based on the prediction custom parameters;

[0007] The predicted state quantity is obtained by processing the vehicle preset parameters and the spatial prediction parameters through the state prediction model.

[0008] Obtain a path reference sequence, and determine the position prediction error based on the path reference sequence and the predicted state quantity;

[0009] Obtain a custom weighting matrix, and determine the control gain based on the custom weighting matrix and the predicted state variables;

[0010] The predictive control increment is determined based on the position prediction error and the control gain;

[0011] Obtain the current control quantity, and determine the target control quantity based on the predicted control increment and the current control quantity.

[0012] A vehicle lateral control device, comprising:

[0013] The prediction model acquisition module is used to acquire vehicle preset parameters, spatial prediction parameters, and prediction custom parameters, and configure the state prediction model according to the prediction custom parameters;

[0014] The module for obtaining predicted state quantities is used to process the vehicle preset parameters and the spatial prediction parameters through the state prediction model to obtain predicted state quantities.

[0015] The location prediction error acquisition module is used to acquire a path reference sequence and determine the location prediction error based on the path reference sequence and the prediction state quantity.

[0016] The control gain acquisition module is used to acquire a custom weighting matrix and determine the control gain based on the custom weighting matrix and the predicted state quantity.

[0017] A predictive control increment acquisition module is used to determine the predictive control increment based on the position prediction error and the control gain;

[0018] The target control quantity acquisition module is used to acquire the current control quantity and determine the target control quantity based on the predicted control increment and the current control quantity.

[0019] A computer device includes a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the above-described vehicle lateral control method when executing the computer-readable instructions.

[0020] One or more readable storage media storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the vehicle lateral control method described above.

[0021] The aforementioned vehicle lateral control method, device, computer equipment, and storage medium use a state prediction model instead of traditional dynamic equations, which can avoid the problem of model inaccuracy caused by system parameter errors and various interference errors, thereby improving the accuracy and stability of vehicle lateral control. At the same time, two adjustable custom parameters are set, which on the one hand allows designers to flexibly adjust the tracking effect; on the other hand, it can achieve a balance between computing power and cost according to hardware performance and control requirements. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention 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.

[0023] Figure 1This is a flowchart illustrating a vehicle lateral control method according to an embodiment of the present invention;

[0024] Figure 2 This is a comparison diagram of the actual output position and the ideal output position under the first condition in one embodiment of the present invention;

[0025] Figure 3 This is one embodiment of the present invention. Figure 2 A magnified view of a section of the comparison image;

[0026] Figure 4 This is a comparison diagram of the actual output curve and the ideal output curve when entering the steady state under the first condition in one embodiment of the present invention;

[0027] Figure 5 This is a comparison diagram of the actual output position and the ideal output position under the second condition in one embodiment of the present invention;

[0028] Figure 6 This is one embodiment of the present invention. Figure 5 A magnified view of a section of the comparison image;

[0029] Figure 7 This is a comparison diagram of the actual output curve and the ideal output curve when entering the steady state under the second condition in one embodiment of the present invention;

[0030] Figure 8 This is a comparison diagram of the actual output position (lateral position) and the ideal output position under the second condition in one embodiment of the present invention;

[0031] Figure 9 This is one embodiment of the present invention. Figure 8 A magnified view of a section of the comparison image;

[0032] Figure 10 This is a comparison diagram of the actual output curve and the ideal output curve when entering the steady state under the second condition in one embodiment of the present invention;

[0033] Figure 11 This is a schematic diagram of a vehicle lateral control device according to an embodiment of the present invention;

[0034] Figure 12 This is a schematic diagram of a computer device according to an embodiment of the present invention. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. 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.

[0036] In one embodiment, such as Figure 1 As shown, a vehicle lateral control method is provided, including the following steps S10 to S60.

[0037] S10. Obtain vehicle preset parameters, spatial prediction parameters, and prediction custom parameters, and configure the state prediction model according to the prediction custom parameters.

[0038] S20. The vehicle preset parameters and the spatial prediction parameters are processed by the state prediction model to obtain the predicted state quantity.

[0039] Understandably, vehicle preset parameters can be measurable parameters of the vehicle used in the state prediction model. Spatial prediction parameters refer to parameters in the state prediction model that cannot be directly measured by the vehicle. Predictive custom parameters refer to the updated parameters of the state prediction model.

[0040] In one example, the custom prediction parameters include prediction step size p and control step size m, which can be set according to the chip's actual computing power and control accuracy requirements.

[0041] The state prediction model can be expressed as:

[0042]

[0043] in,

[0044] Y p (k+1|k) represents the predicted state quantity at time k+1;

[0045] y c (k+1|k) is the state variable of the controlled output at time k+1;

[0046] y c (k+2|k) is the state variable of the controlled output at time k+2;

[0047] y c (k+p|k) is the state variable of the controlled output at time k+p;

[0048]

[0049] I is the element matrix;

[0050] The output of the state observer at time k is the spatial prediction parameter.

[0051] S u The system unit step response coefficient matrix with zero initial state;

[0052] ΔU(k) ​​is the change in the input;

[0053] Δu(k) is the change in input of the step response at time k;

[0054] Δu(k+1) is the change in input of the step response at time k+1;

[0055] Δu(k+m-1) is the change in input of the step response at time k+m-1;

[0056]

[0057] M ss-iteration This is the expression for the free response output matrix of the state prediction model under zero input.

[0058] C c This is the controlled output coefficient matrix, which is the coefficient matrix of the system output and the controlled output;

[0059] C is the coefficient matrix of the state and current system output of the unit step response model;

[0060] C m The measurement output coefficient matrix is ​​a matrix of coefficients between the system output and the measurement output.

[0061] M ss The system's free response output matrix under zero input;

[0062]

[0063] S u-iteration This is the system unit step response coefficient matrix with zero initial state in the state prediction model;

[0064]

[0065] S d-iteration This represents the unit step response coefficient matrix from the initial disturbance to the output of the state prediction model.

[0066] S d The unit step response coefficient matrix from the disturbance with zero initial state to the output.

[0067] It should be noted that the system mentioned above refers to the vehicle state step model.

[0068] S30. Obtain the path reference sequence, and determine the position prediction error based on the path reference sequence and the predicted state quantity.

[0069] Understandably, the path reference sequence can be represented as:

[0070]

[0071] Where R(k+1) is the target's lateral position at time k+1;

[0072] r(k+1) is the path reference value at time k+1;

[0073] r(k+2) is the path reference value at time k+2;

[0074] r(k+p) is the path reference value at time k+p.

[0075] The location prediction error can be calculated using the following formula:

[0076]

[0077] Among them, E p (k+1|k) represents the position prediction error, which is the estimated value of the lateral position error from the current time to the next time.

[0078] R(k+1) represents the horizontal position of the target at time k+1;

[0079] M ss-iteration This is the expression for the free response output matrix of the state prediction model under zero input.

[0080] The output of the state observer at time k is the spatial prediction parameter.

[0081] S d-iteration This represents the unit step response coefficient matrix from the initial disturbance to the output of the state prediction model.

[0082] Δd(k) represents the change in system disturbance at time k.

[0083] S40. Obtain a custom weighting matrix, and determine the control gain based on the custom weighting matrix and the predicted state variables.

[0084] Understandably, a custom weighting matrix includes a process cost weighting matrix and a control cost weighting matrix. Custom weighting matrices can be set according to control objectives and system stability.

[0085] In one example, the cost function can be constructed as follows:

[0086]

[0087] J is the value of the cost function;

[0088] Γ y,i This refers to the process cost weighting coefficient.

[0089] Γ u,i To control the cost weighting coefficient;

[0090] p is the prediction step size;

[0091] m is the control step size;

[0092] y c (k+i|k) is the state variable of the controlled output at time k+i;

[0093] r(k+i) is the path reference value at time k+i;

[0094] Δu(k+i-1) is the change in input of the step response at time k+i-1.

[0095] After rewriting the cost function in vector form, a corresponding custom weighting matrix can be set. Specifically:

[0096] Γ y =diag{Γ y,i}| i＝1～p

[0097] Γ u =diag{Γ u,i}| i＝1～m

[0098] Γ y This is the process cost weight matrix;

[0099] Γ u To control the cost weight matrix;

[0100] diag{} is the diag function, used to construct a diagonal matrix;

[0101] p is the prediction step size;

[0102] m is the control step size.

[0103] The control gain can be determined based on a custom weighting matrix and the predicted state variables. In one example, the formula for determining the control gain can be expressed as:

[0104]

[0105] Among them, K mpc Indicates control gain;

[0106] [I 0…0] 1*m Let I represent a 1*m dimensional row matrix, where I is the identity matrix;

[0107] S represents u-iteration The transpose of the matrix;

[0108] This is the transpose of the process cost weight matrix;

[0109] Γ y This is the process cost weight matrix;

[0110] S u-iteration This is the system unit step response coefficient matrix with zero initial state in the state prediction model;

[0111] To control the transpose of the cost weight matrix;

[0112] Γ u To control the cost weight matrix.

[0113] In some examples, the stability of a closed-loop system is determined using the system matrix. The system matrix is ​​represented as:

[0114] M ss -S u-iteration K mpc M ss -iteration

[0115] M ss The system's free response output matrix under zero input;

[0116] S u-iteration This is the system unit step response coefficient matrix with zero initial state in the state prediction model;

[0117] K mpc Indicates control gain;

[0118] M ss-iteration This represents the expression for the free response output matrix of the state prediction model under zero input.

[0119] S50. Determine the predictive control increment based on the position prediction error and the control gain;

[0120] S60. Obtain the current control quantity, and determine the target control quantity based on the predicted control increment and the current control quantity.

[0121] Understandably, the predictive control increment can be determined from the position prediction error and the control gain. In one example, the predictive control increment can be calculated using the following formula:

[0122] Δu(k)=K mpc E p (k+1|k)

[0123] Where Δu(k) represents the predictive control increment at time k;

[0124] K mpc Indicates control gain;

[0125] Ep (k+1|k) represents the position prediction error at time k+1.

[0126] The current control variable u(k-1) can be obtained, and by superimposing it with Δu(k), the target control variable u(k) can be obtained. That is:

[0127] u(k) = u(k-1) + Δu(k)

[0128] Here, the target control quantity u(k) is the control quantity applied to the system at time k+1.

[0129] In this embodiment, a state prediction model is used instead of the traditional dynamic equations, which can avoid the problem of model inaccuracy caused by system parameter errors and various interference errors, and improve the accuracy and stability of vehicle lateral control. At the same time, two adjustable custom parameters are set, which can set the accuracy of the model according to actual needs. On the one hand, it allows designers to flexibly adjust the tracking effect; on the other hand, it can achieve a balance between computing power and cost according to hardware performance and control requirements.

[0130] Optionally, step S10, namely obtaining the spatial prediction parameters, includes:

[0131] S101, Obtain the state observer;

[0132] S102. Determine whether the state observer is stable;

[0133] S103. If the state observer is stable, then predict the spatial prediction parameters based on the state observer.

[0134] Understandably, the state observer is used to predict parameters in the state prediction model that the vehicle cannot directly measure, i.e., to obtain spatial prediction parameters.

[0135] In one example, the state observer can be set as follows:

[0136]

[0137] in, This is the output of the state observer at time k;

[0138] M ss The system's free response output matrix under zero input;

[0139] This is the output of the state observer at time k-1;

[0140] S u The system unit step response coefficient matrix with zero initial state;

[0141] Δu(k-1) is the change in input of the step response at time k-1;

[0142] S d The unit step response coefficient matrix from the disturbance with zero initial state to the output;

[0143] Δd(k-1) represents the change in system disturbance at time k-1;

[0144] K F For custom observer matrices;

[0145] y m For measurement output;

[0146] C m To measure the output coefficient matrix;

[0147] C is the coefficient matrix of the state of the unit step response model and the current system output.

[0148] Before using the state observer, it is necessary to determine whether the state observer is stable. If the state observer is stable, the spatial prediction parameters can be predicted based on it. If the state observer is unstable, the custom observer matrix K needs to be reset. F This continues until the state observer stabilizes. In other words, the observer matrix K needs to be designed based on the stability of the state observer. F .

[0149] In this embodiment, the stability of spatial prediction parameters can be ensured by selecting a state observer based on stability.

[0150] Optionally, before step S101, i.e. before acquiring the state observer, the following steps are included:

[0151] S110. Obtain the vehicle state step model, the controlled output coefficient matrix, and the measured output coefficient matrix;

[0152] S111. Evaluate the observability of the vehicle state step model based on the controlled output coefficient matrix and the measured output coefficient matrix;

[0153] S112. If the vehicle state step model is observable, then configure the state observer according to the controlled output coefficient matrix, the measurement, and the vehicle state step model.

[0154] Understandably, a vehicle state step model can be a vehicle space state model constructed based on a step response. The controlled output coefficient matrix is ​​the coefficient matrix between the system output and the controlled output. The measured output coefficient matrix is ​​the coefficient matrix between the system output and the measured output. The controlled output matrix C can be set according to the actual situation. c and measurement output matrix C mThe measurement output matrix C can be designed according to the observer's stabilization requirements. m The controlled output matrix C can be designed independently based on the variables of interest. c .

[0155] In one example, the controlled output coefficient matrix and the measured output coefficient matrix can be represented as:

[0156] y(k)=CY(k)

[0157] y c (k)=C c y(k)

[0158] y m (k)=C m y(k)

[0159] Where y(k) is the lateral position of the vehicle at time k;

[0160] C is the coefficient matrix of the state and current system output of the unit step response model;

[0161] Y(k) is the output of the step response at time k;

[0162] y c (k) represents the state variable of the controlled output at time k;

[0163] C c This is the controlled output coefficient matrix, which is the coefficient matrix of the system output and the controlled output;

[0164] y m (k) represents the measurement output at time k;

[0165] C m The measurement output coefficient matrix is ​​a matrix of coefficients between the system output and the measurement output.

[0166] The observability of the vehicle state step model can be evaluated using the following equation:

[0167]

[0168] Among them, C m The measurement output coefficient matrix is ​​a matrix of coefficients between the system output and the measurement output.

[0169] C is the coefficient matrix of the state and current system output of the unit step response model;

[0170] M ss M is the system free response output matrix under zero input; ss It is an N*N matrix.

[0171] If the rank matrix above is full rank, then the system is observable.

[0172] This embodiment achieves predictability of spatial prediction parameters by evaluating the observability of the vehicle state step model.

[0173] Optionally, after step S111, that is, after evaluating the observability of the vehicle state step model based on the controlled output coefficient matrix and the measured output coefficient matrix, the method further includes:

[0174] S113. If the vehicle state step model is not observable, then receive a matrix modification instruction.

[0175] S114. Modify the measurement output coefficient matrix according to the matrix modification instruction, so as to evaluate the observability of the vehicle state step model based on the controlled output coefficient matrix and the modified measurement output coefficient matrix.

[0176] Understandably, if the vehicle state step model is not observable, the measurement output coefficient matrix needs to be modified using matrix modification instructions until the vehicle state step model becomes observable.

[0177] This embodiment modifies the measurement output coefficient matrix to make the vehicle state step model observable.

[0178] Optionally, before step S110, i.e. before obtaining the vehicle state step model, the method further includes:

[0179] S120, Obtain the first step model from control to output, and the second step model from measurable disturbance to output;

[0180] S121. Superimpose the first step model and the second step model to obtain the first vehicle state step model;

[0181] S122. Perform accuracy verification on the first vehicle state step model;

[0182] S123. If the first vehicle state step model passes the accuracy verification, then the first vehicle state step model is determined as the vehicle state step model.

[0183] Understandably, a state-space unit step response model from control to output can be established, which is the first step model. In one example, the first step model can be represented as:

[0184] Y(k)=M ss Y(k-1)+S u Δu(k-1)

[0185] Where Y(k) represents the output of the step response at time k;

[0186] M ss The system's free response output matrix under zero input;

[0187] Y(k-1) represents the output of the step response at time k-1;

[0188] S u The system unit step response coefficient matrix with zero initial state;

[0189] Δu(k-1) is the change in input of the step response at time k-1.

[0190] In the first step model, the first term is the free response of the system with zero input; the second term is the forced response with zero coefficients.

[0191] For systems where there is no integral element in the transfer function matrix (see step S131):

[0192]

[0193]

[0194] Among them, M ss The system's free response output matrix under zero input;

[0195] For n y *n y An identity matrix of order 1;

[0196] n y The step size for the output;

[0197] N is the total number of sampling points;

[0198] S u The system unit step response coefficient matrix with zero initial state;

[0199] S1 is the response value of the first unit step response point under control only;

[0200] S N This is the response value of the Nth unit step response point under control only;

[0201] s ij,k To control the response value of the k-th unit step response point when the input step size is i and the output step size is j.

[0202] A state-space unit step response model to a measurable disturbance can be established, which is called the second step model. In one example, the second step model can be expressed as:

[0203] Y(k)=M ssY(k-1)+S d Δd(k-1)

[0204] Where Y(k) represents the output of the step response at time k;

[0205] M ss The system's free response output matrix under zero input;

[0206] Y(k-1) represents the output of the step response at time k-1;

[0207] S d The unit step response coefficient matrix from the disturbance with zero initial state to the output;

[0208] Δd(k-1) represents the change in system disturbance at time k-1.

[0209] In the second step model, the first term is the free response of the system with zero input; the second term is the forced response with zero coefficients.

[0210] For systems where the transfer function matrix does not contain an integral element:

[0211]

[0212]

[0213] Where S1 is the response value of the first unit step response point under the influence of only disturbance;

[0214] S N This represents the response value at the Nth unit step response point under interference only.

[0215] s ij,k This is the response value at the k-th unit step response point when the input step size is i and the output step size is j.

[0216] The definitions of other parameters are the same as those in the first step model.

[0217] By superimposing the first and second step models, a first vehicle state step model can be obtained. Then, the accuracy of the first vehicle state step model is verified. If the first vehicle state step model passes the accuracy verification, it is confirmed as a vehicle state step model. If the first vehicle state step model fails the accuracy verification, the sampling time and steady-state error determination conditions need to be readjusted.

[0218] In one example, the vehicle state step model can be represented as:

[0219] Y(k)=M ss Y(k-1)+S uΔu(k-1)+S d Δd(k-1)

[0220] Where Y(k) is the output of the step response at time k;

[0221] M ss The system's free response output matrix under zero input;

[0222] Y(k-1) represents the output of the step response at time k-1;

[0223] S u The system unit step response coefficient matrix with zero initial state;

[0224] Δu(k-1) is the change in input of the step response at time k-1;

[0225] S d The unit step response coefficient matrix from the disturbance with zero initial state to the output.

[0226] This embodiment obtains a vehicle state step model that meets the accuracy requirements through accuracy verification.

[0227] Optionally, before step S120, that is, before obtaining the first step model from control to output and the second step model from measurable disturbance to output, the method further includes:

[0228] S130. Construct the two-degree-of-freedom dynamic state-space equations for the vehicle;

[0229] S131. Transform the two-degree-of-freedom dynamic state-space equations of the vehicle to obtain the transfer function matrix;

[0230] S132. Obtain vehicle system parameters and custom parameters for spatial status;

[0231] S133. Configure the first step model and the second step model according to the transfer function matrix, the vehicle system parameters and the space state custom parameters.

[0232] Understandably, the two-degree-of-freedom dynamic state-space equations of a vehicle can be expressed as:

[0233]

[0234] in, Indicates the lateral acceleration of the vehicle;

[0235] This indicates the vehicle's yaw acceleration;

[0236] Indicates the lateral speed of the vehicle;

[0237] Indicates the vehicle's yaw rate;

[0238] Indicates the front wheel lateral stiffness;

[0239] Indicates the rear wheel lateral stiffness;

[0240] l f This indicates the distance from the center of gravity to the front axle;

[0241] l r This indicates the distance from the center of mass to the rear axle;

[0242] v x Indicates the longitudinal speed of the vehicle;

[0243] m represents the total mass of the vehicle;

[0244] I z This represents the moment of inertia of the vehicle about the z-axis.

[0245] δ represents the front wheel steering angle.

[0246] The transfer function matrix of the MIMO (Multiple Input Multiple Output) system is obtained by transforming the state-space equations of the vehicle's two degrees of freedom dynamics.

[0247] In one example, the transfer function matrix can be represented as:

[0248] G(s)=C(λI-A) -1 BD

[0249] G(s) represents the state-space equation of the system;

[0250] A represents the state matrix;

[0251] B represents the system's control matrix;

[0252] C represents the system's output matrix;

[0253] D represents the direct transfer matrix of the system;

[0254] λ represents the eigenvalues ​​of matrix A;

[0255] I represents the identity matrix.

[0256] A unit step input to the vehicle steering wheel can be made to collect vehicle system parameters. These parameters include the vehicle's lateral position (e.g., y(k)~y(k+N-1), the vehicle's lateral position from time k to k+N-1), and lateral velocity (e.g., ...). Lateral velocity from k to k+N-1, heading angle (e.g.) Vehicle yaw angle from k to k+N-1), yaw rate output (e.g., vehicle ... (vehicle yaw rate from k to k+N-1).

[0257] The custom parameters for spatial state include the steady-state error band σ and the sampling time Ts.

[0258] The first and second step models can be configured by combining the transfer function matrix, vehicle system parameters, and custom parameters of spatial state. Then, the first and second step models can be superimposed to obtain the vehicle state step model.

[0259] This embodiment utilizes a step input to configure the first and second step models, which can avoid the problem of model inaccuracy caused by system parameter errors and various interference errors.

[0260] Optionally, after step S122, i.e. after performing accuracy verification on the first vehicle state step model, the process includes:

[0261] S124. If the first vehicle state step model fails the accuracy verification, a parameter modification instruction is received, and the spatial state custom parameters are modified according to the parameter modification instruction to obtain the second spatial state custom parameters, so as to generate the second vehicle state step model to be verified according to the second spatial state custom parameters.

[0262] Understandably, if the first vehicle state step model fails the accuracy check, the custom parameters of the spatial state are modified using a parameter modification command to obtain the custom parameters of the second spatial state. Then, a second vehicle state step model to be checked is generated based on the second custom parameters of the second spatial state. The accuracy check of the second vehicle state step model continues until the Nth vehicle state step model passes the accuracy check.

[0263] In one application example, the initial planned lateral position is 1, and the vehicle's initial lateral position is 0. Combined with... Figures 2-4 It can be seen that under the first condition (the system has no measurable or unmeasurable errors), the actual output can smoothly follow the target trajectory in just 0.17 seconds, with a maximum error of 0.163m. Figure 2 This is a comparison chart of the actual output position (horizontal position) and the ideal output position under the first condition. Figure 3 for Figure 2 A magnified view of a portion of the comparison image. Figure 4 This is a comparison chart of the actual output curve and the ideal output curve when entering steady state under the first condition.

[0264] Combination Figures 5-7 It can be seen that in the second situation (the system has a measurable error of 0.1m and the control parameters have not been adjusted in any way), the actual output can still keep up with the target trajectory smoothly in just 0.21 seconds, while the maximum error is 0.172m. Figure 5 This is a comparison chart of the actual output position (horizontal position) and the ideal output position under the second condition. Figure 6 for Figure 5 A magnified view of a portion of the comparison image. Figure 7 This is a comparison chart of the actual output curve and the ideal output curve when entering steady state under the second condition.

[0265] Combination Figures 8-10 It can be seen that under the third condition (the system has a measurable error of 0.1m and an unmeasurable error of 0.1m, and the control parameters have not been adjusted in any way), the actual output can still keep up with the target trajectory smoothly, enter the 2% steady-state error band in 0.78 seconds, and the maximum error is 0.27m. Figure 8 This is a comparison chart of the actual output position (horizontal position) and the ideal output position under the second condition. Figure 9 for Figure 8 A magnified view of a portion of the comparison image. Figure 10 This is a comparison chart of the actual output curve and the ideal output curve when entering steady state under the second condition.

[0266] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0267] In one embodiment, a vehicle lateral control device is provided, which corresponds one-to-one with the vehicle lateral control method described in the above embodiments. For example... Figure 11 As shown, the vehicle lateral control device includes a prediction model acquisition module 10, a prediction state quantity acquisition module 20, a position prediction error acquisition module 30, a control gain acquisition module 40, a prediction control increment acquisition module 50, and a target control quantity acquisition module 60. Detailed descriptions of each functional module are as follows:

[0268] The prediction model acquisition module 10 is used to acquire vehicle preset parameters, spatial prediction parameters and prediction custom parameters, and configure the state prediction model according to the prediction custom parameters.

[0269] The module 20 for obtaining predicted state quantities is used to process the vehicle preset parameters and the spatial prediction parameters through the state prediction model to obtain predicted state quantities.

[0270] The location prediction error acquisition module 30 is used to acquire a path reference sequence and determine the location prediction error based on the path reference sequence and the prediction state quantity.

[0271] The control gain acquisition module 40 is used to acquire a custom weighting matrix and determine the control gain based on the custom weighting matrix and the predicted state quantity.

[0272] A predictive control increment module 50 is used to determine the predictive control increment based on the position prediction error and the control gain.

[0273] The target control quantity acquisition module 60 is used to acquire the current control quantity and determine the target control quantity based on the predicted control increment and the current control quantity.

[0274] Optionally, the prediction model module 10 includes:

[0275] The state observer acquisition unit is used to acquire the state observer;

[0276] A stability determination unit is used to determine whether the state observer is stable;

[0277] The prediction parameter unit is used to predict the spatial prediction parameters based on the state observer if the state observer is stable.

[0278] Optionally, the vehicle lateral control device further includes a configuration observer module, the configuration observer module comprising:

[0279] The model and matrix unit are used to obtain the vehicle state step model, the controlled output coefficient matrix, and the measured output coefficient matrix.

[0280] An observability evaluation unit is used to evaluate the observability of the vehicle state step model based on the controlled output coefficient matrix and the measured output coefficient matrix.

[0281] A configuration observer unit is configured to configure the state observer based on the controlled output coefficient matrix, the measurement, and the vehicle state step model if the vehicle state step model is observable.

[0282] Optionally, the configuration observer module further includes:

[0283] The receiving matrix modification instruction unit is used to receive a matrix modification instruction if the vehicle state step model is not observable.

[0284] A matrix modification unit is used to modify the measurement output coefficient matrix according to the matrix modification instruction, so as to evaluate the observability of the vehicle state step model based on the controlled output coefficient matrix and the modified measurement output coefficient matrix.

[0285] Optionally, the vehicle lateral control device further includes a step model configuration module, the step model configuration module comprising:

[0286] First and second step model units are obtained to acquire the first step model from control to output and the second step model from measurable disturbance to output.

[0287] The superposition unit is used to superimpose the first step model and the second step model to obtain the first vehicle state step model;

[0288] The accuracy verification unit is used to verify the accuracy of the first vehicle state step model.

[0289] A step model unit is obtained, which is used to determine the first vehicle state step model as the vehicle state step model if the first vehicle state step model passes the accuracy verification.

[0290] Optionally, the step model configuration module also includes:

[0291] Construct state equation units to build the two-degree-of-freedom dynamic state-space equations of the vehicle;

[0292] The transformation unit is used to transform the two-degree-of-freedom dynamic state-space equations of the vehicle to obtain the transfer function matrix;

[0293] The parameter acquisition unit is used to acquire vehicle system parameters and custom parameters of spatial status;

[0294] A configuration model unit is used to configure the first step model and the second step model according to the transfer function matrix, the vehicle system parameters and the space state custom parameters.

[0295] Optionally, the step model configuration module also includes:

[0296] The parameter modification unit is used to receive a parameter modification instruction if the first vehicle state step model fails the accuracy verification, modify the spatial state custom parameters according to the parameter modification instruction, obtain the second spatial state custom parameters, and generate a second vehicle state step model to be verified according to the second spatial state custom parameters.

[0297] Specific limitations regarding the vehicle lateral control device can be found in the limitations of the vehicle lateral control method described above, and will not be repeated here. Each module in the aforementioned vehicle lateral control device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in hardware or independently of the processor in a computer device, or stored in software in the memory of a computer device, so that the processor can call and execute the corresponding operations of each module.

[0298] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 12As shown, the computer device includes a processor, memory, network interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a readable storage medium and internal memory. The non-volatile storage medium stores an operating system and computer-readable instructions. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in the readable storage medium. The network interface is used to communicate with an external server via a network connection. When the computer-readable instructions are executed by the processor, they implement a vehicle lateral control method. The readable storage medium provided in this embodiment includes both non-volatile and volatile readable storage media.

[0299] In one embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor performs the following steps when executing the computer-readable instructions:

[0300] Obtain vehicle preset parameters, spatial prediction parameters, and prediction custom parameters, and configure a state prediction model based on the prediction custom parameters;

[0301] The predicted state quantity is obtained by processing the vehicle preset parameters and the spatial prediction parameters through the state prediction model.

[0302] Obtain a path reference sequence, and determine the position prediction error based on the path reference sequence and the predicted state quantity;

[0303] Obtain a custom weighting matrix, and determine the control gain based on the custom weighting matrix and the predicted state variables;

[0304] The predictive control increment is determined based on the position prediction error and the control gain;

[0305] Obtain the current control quantity, and determine the target control quantity based on the predicted control increment and the current control quantity.

[0306] In one embodiment, one or more computer-readable storage media storing computer-readable instructions are provided. The readable storage media provided in this embodiment include non-volatile readable storage media and volatile readable storage media. The readable storage media stores computer-readable instructions, which, when executed by one or more processors, perform the following steps:

[0307] Obtain vehicle preset parameters, spatial prediction parameters, and prediction custom parameters, and configure a state prediction model based on the prediction custom parameters;

[0308] The predicted state quantity is obtained by processing the vehicle preset parameters and the spatial prediction parameters through the state prediction model.

[0309] Obtain a path reference sequence, and determine the position prediction error based on the path reference sequence and the predicted state quantity;

[0310] Obtain a custom weighting matrix, and determine the control gain based on the custom weighting matrix and the predicted state variables;

[0311] The predictive control increment is determined based on the position prediction error and the control gain;

[0312] Obtain the current control quantity, and determine the target control quantity based on the predicted control increment and the current control quantity.

[0313] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware with computer-readable instructions. These computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When executed, these computer-readable instructions can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0314] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0315] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A vehicle lateral control method characterized by, The method comprises: acquiring vehicle preset parameters, space prediction parameters and prediction custom parameters, and configuring a state prediction model according to the prediction custom parameters; processing the vehicle preset parameters and the space prediction parameters through the state prediction model to obtain a prediction state quantity; acquiring a path reference sequence, and determining a position prediction error according to the path reference sequence and the prediction state quantity; acquiring a custom weighting matrix, and determining a control gain according to the custom weighting matrix and the prediction state quantity; determining a prediction control increment according to the position prediction error and the control gain; acquiring a current control quantity, and determining a target control quantity according to the prediction control increment and the current control quantity.

2. The vehicle lateral control method according to claim 1, characterized by, The acquiring of the space prediction parameters comprises: acquiring a state observer; determining whether the state observer is stable; if the state observer is stable, predicting the space prediction parameters according to the state observer.

3. The vehicle lateral control method according to claim 2, characterized by, Before the acquiring of the state observer, the method comprises: acquiring a vehicle state step model, a controlled output coefficient matrix and a measurement output coefficient matrix; evaluating the observability of the vehicle state step model according to the controlled output coefficient matrix and the measurement output coefficient matrix; if the vehicle state step model has observability, configuring the state observer according to the controlled output coefficient matrix, the measurement output coefficient matrix and the vehicle state step model.

4. The vehicle lateral control method according to claim 3, characterized by, After the evaluating of the observability of the vehicle state step model according to the controlled output coefficient matrix and the measurement output coefficient matrix, the method further comprises: if the vehicle state step model does not have observability, receiving a matrix modification instruction; modifying the measurement output coefficient matrix according to the matrix modification instruction, so as to evaluate the observability of the vehicle state step model according to the controlled output coefficient matrix and the modified measurement output coefficient matrix.

5. The vehicle lateral control method according to claim 3, characterized by, Before the acquiring of the vehicle state step model, the method further comprises: acquiring a first step model from control to output and a second step model from measurable disturbance to output; superimposing the first step model and the second step model to obtain a first vehicle state step model; performing precision verification on the first vehicle state step model; if the first vehicle state step model passes the precision verification, determining the first vehicle state step model as the vehicle state step model.

6. The vehicle lateral control method according to claim 5, characterized by, Before the acquiring of the first step model from control to output and the second step model from measurable disturbance to output, the method further comprises: constructing a vehicle two-degree-of-freedom dynamics state space equation; transforming the vehicle two-degree-of-freedom dynamics state space equation to obtain a transmission function matrix; acquiring vehicle system parameters and space state custom parameters; configuring the first step model and the second step model according to the transmission function matrix, the vehicle system parameters and the space state custom parameters.

7. The vehicle lateral control method according to claim 6, characterized by, After the performing of the precision verification on the first vehicle state step model, the method comprises: If the first vehicle state step model fails the accuracy check, a parameter modification instruction is received, the spatial state custom parameter is modified according to the parameter modification instruction, a second spatial state custom parameter is obtained, and a second vehicle state step model to be checked is generated according to the second spatial state custom parameter.

8. A vehicle lateral control device characterized by comprising: Comprise: An acquisition of a prediction model module is configured to acquire vehicle preset parameters, spatial prediction parameters, and prediction custom parameters, and configure a state prediction model according to the prediction custom parameters; An acquisition of a prediction state quantity module is configured to process the vehicle preset parameters and the spatial prediction parameters through the state prediction model to obtain a prediction state quantity; An acquisition of a position prediction error module is configured to acquire a path reference sequence, and determine a position prediction error according to the path reference sequence and the prediction state quantity; An acquisition of a control gain module is configured to acquire a custom weighting matrix, and determine a control gain according to the custom weighting matrix and the prediction state quantity; An acquisition of a prediction control increment module is configured to determine a prediction control increment according to the position prediction error and the control gain; An acquisition of a target control quantity module is configured to acquire a current control quantity, and determine a target control quantity according to the prediction control increment and the current control quantity.

9. A computer device comprising a memory, a processor, and computer readable instructions stored in the memory and executable on the processor, wherein, The processor implements the vehicle lateral control method in any one of claims 1 to 7 when executing the computer readable instructions.

10. One or more readable storage media storing computer readable instructions, which are executed by one or more processors to cause the one or more processors to implement the vehicle lateral control method in any one of claims 1 to 7.

Images