Vehicle control method and apparatus, and storage medium, electronic device and vehicle

Abstract

A vehicle control method, comprising: determining, as first pose information, pose information of a vehicle at a target moment and pose information respectively corresponding to each prediction point (11); determining a plurality of reference points on a reference travel path corresponding to the vehicle, and determining, as second pose information, pose information respectively corresponding to each reference point (12); on the basis of the first pose information and the second pose information, constructing an LQR optimization model (13); on the basis of the LQR optimization model, determining target control parameters at the next moment of the target moment (14); and at the next moment of the target moment, controlling the vehicle on the basis of the target control parameters (15).

Description

Vehicle control method, device, storage medium, electronic device and vehicle

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This disclosure claims priority to Chinese patent application number 202311440358.1, filed with the China Patent Office on October 31, 2023, entitled “Vehicle Control Method, Device, Storage Medium, Electronic Device and Vehicle,” the entire contents of which are incorporated by reference into this disclosure. Technical Field

[0003] The present disclosure relates to the field of vehicle control, and in particular, to a vehicle control method, device, storage medium, electronic device, and vehicle. Background Art

[0004] In intelligent driving scenarios, to prevent vehicles from deviating from their lanes or colliding with obstacles, intelligent driving control systems generally have high requirements for lateral control of the vehicle. Currently, commonly used control methods include PID (Proportion Integral Differential) control, pure tracking control, LQR (Linear Quadratic Regulator) control, and MPC (Model Prediction Control). PID control and pure tracking control have low precision, and MPC consumes more computing power, making LQR more commonly used. However, related LQR-based control schemes suffer from control instability.

[0005] Summary of the Invention

[0006] The present invention aims to provide a vehicle control method, device, storage medium, electronic device and vehicle to improve the control accuracy and stability of the vehicle.

[0007] To achieve the above objectives, according to a first aspect of the present disclosure, a vehicle control method is provided, the method comprising:

[0008] Determining the vehicle's posture information at a target time and the posture information corresponding to each of a plurality of prediction points as first posture information, wherein the prediction points are determined from the predicted path corresponding to the vehicle;

[0009] Determining a plurality of reference points on a reference driving path corresponding to the vehicle, and determining pose information corresponding to each reference point as second pose information, the plurality of reference points including a first reference point corresponding to the target time and a preset number of second reference points after the target time, and each of the reference points having a corresponding prediction point on the predicted path;

[0010] Constructing an LQR optimization model according to the first posture information and the second posture information;

[0011] Determining target control parameters at a next moment of the target moment according to the LQR optimization model;

[0012] At a moment next to the target moment, the vehicle is controlled according to the target control parameter.

[0013] Optionally, the position information of the vehicle at the target time includes the coordinates of the vehicle at the target time;

[0014] Determining a plurality of reference points on the reference driving path of the vehicle includes:

[0015] Determine a projection point of the coordinates of the vehicle at the target time on the reference driving path as the first reference point;

[0016] Taking the location of the first reference point as the starting location and the preset distance as the spacing between adjacent reference points, the preset number of second reference points are determined one by one on the reference driving path.

[0017] Optionally, the reference driving path corresponds to a path equation, and the posture information includes coordinates and heading angles;

[0018] Determining the pose information corresponding to each reference point includes:

[0019] Determining the coordinates corresponding to each of the reference points according to the reference driving path;

[0020] For each reference point, the heading angle corresponding to the reference point is determined according to the coordinates corresponding to the reference point and the path equation.

[0021] Optionally, the heading angle θ corresponding to the i-th reference point Ri is determined by the following formula: Ri :

[0022] θ Ri =arctan(f′(x Ri ))

[0023] Wherein, 0≤i≤preset number, R0 is the first reference point, f(x Ri ) is the path equation, f′(x Ri ) is the derivative of the path equation.

[0024] Optionally, the LQR optimization model includes an objective cost function;

[0025] The target cost function is determined by:

[0026] Get the pose cost coefficient and control cost coefficient;

[0027] Generate a first cost function according to the first pose information, the second pose information and the pose cost coefficient;

[0028] generating a second cost function according to the preset control parameters and the control cost coefficient;

[0029] The sum of the first cost function and the second cost function is determined as the target cost function.

[0030] Optionally, the target cost function F is generated according to the following formula:

[0031] in, is the first cost function, is the second cost function; X i is the pose information predicted for the i-th prediction point, X Ri is the second pose information of the i-th reference point, X R0 is the second pose information of the first reference point, N is the preset number, Q is the diagonal matrix generated according to the pose cost coefficient; r is the control cost coefficient, κ i is the control parameter of the i-th prediction point.

[0032] Optionally, the posture information includes coordinates and heading angles, and the posture cost coefficient includes a position cost coefficient and a heading cost coefficient;

[0033] Q is generated according to the following formula:

[0034] Q = diag[q xy ,q xy ,q θ ]

[0035] Among them, q xy is the position cost coefficient, q θ is the heading cost coefficient.

[0036] Optionally, the LQR optimization model further includes a target constraint condition, wherein the target constraint condition includes a first constraint condition for constraining a generation method of the predicted path and a second constraint condition for constraining X0 to be the first posture information.

[0037] Optionally, the posture information includes coordinates and heading angles;

[0038] The first constraint condition is generated in the following way:

[0039] Construct a kinematic model with posture information as state quantity and preset control parameters as control quantity;

[0040] Discretizing the kinematic model according to a preset distance to obtain a system equation for generating position information of a second prediction point according to the position information of the first prediction point, where the second prediction point is a next prediction point separated from the first prediction point by a preset distance;

[0041] The system equation is linearized to obtain a processed target equation as the first constraint condition.

[0042] Optionally, the system equation is:

[0043] Among them, X i is the predicted pose information for the i-th prediction point, Δs is the preset distance, θ i is the heading angle corresponding to the i-th predicted point.

[0044] Optionally, the linearization process is Taylor expansion;

[0045] The objective equation is:

[0046] X i+1 =A i X i +B i κ i +C i

[0047] Among them, A i 、B i 、C i Determined by:

[0048] Among them, θ Ri is the heading angle corresponding to the i-th reference point.

[0049] Optionally, determining the target control parameter at the next moment after the target moment according to the LQR optimization model includes:

[0050] Determining a feedback gain matrix that can minimize the target cost function while satisfying the target constraint condition;

[0051] Determine a target control parameter at a next moment of the target moment according to the first posture information and the feedback gain matrix.

[0052] Optionally, determining a feedback gain matrix that enables the target cost function to obtain a minimum value while satisfying the target constraint condition includes:

[0053] The target cost function is solved by using a dynamic programming method to obtain the feedback gain matrix.

[0054] Optionally, determining a target control parameter at a next moment of the target moment according to the first posture information and the feedback gain matrix includes:

[0055] The target control parameter κ is determined according to the following formula:

[0056] κ=-K i X i -L i

[0057] Among them, K i , L i is the feedback gain matrix, and:

[0058] q i =-QX Ri

[0059] Among them, P N =Q,F N =q N .

[0060] Optionally, the target control parameter is steering curvature;

[0061] The controlling the vehicle according to the target control parameter includes:

[0062] determining a front wheel steering angle corresponding to the steering curvature;

[0063] The vehicle is controlled according to the front wheel steering angle.

[0064] According to a second aspect of the present disclosure, there is provided a vehicle control device, the device comprising:

[0065] A first determination module is configured to determine the posture information of the vehicle at a target time and the posture information corresponding to each of a plurality of prediction points as first posture information, wherein the prediction points are determined from the predicted path corresponding to the vehicle;

[0066] a second determining module, configured to determine a plurality of reference points on a reference driving path corresponding to the vehicle, and determine pose information corresponding to each reference point as second pose information, wherein the plurality of reference points include a first reference point corresponding to the target moment and a preset number of second reference points after the target moment, and each of the reference points has a corresponding prediction point on the predicted path;

[0067] A construction module, configured to construct an LQR optimization model based on the first posture information and the second posture information;

[0068] A third determining module is configured to determine a target control parameter at a next moment after the target moment according to the LQR optimization model;

[0069] A control module is used to control the vehicle according to the target control parameter at a moment next to the target moment.

[0070] According to a third aspect of the present disclosure, a non-transitory computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the steps of the vehicle control method described in the first aspect of the present disclosure are implemented.

[0071] According to a fourth aspect of the present disclosure, there is provided an electronic device, including:

[0072] a memory having a computer program stored thereon;

[0073] A processor is used to execute the computer program in the memory to implement the steps of the vehicle control method described in the first aspect of the present disclosure.

[0074] According to a fifth aspect of the present disclosure, a vehicle is provided, comprising the electronic device described in the fourth aspect of the present disclosure.

[0075] Through the above technical solution, the vehicle's posture information at the target moment and the posture information corresponding to multiple predicted points on the predicted path are determined as the first posture information. Multiple reference points and their respective second posture information are determined on the reference driving path corresponding to the vehicle. Based on the first and second posture information, an LQR optimization model is constructed to determine target control parameters for the moment immediately following the target moment, and the vehicle is then controlled according to the target control parameters at the moment immediately following the target moment. The reference points determined on the reference driving path include both the first reference point corresponding to the target moment and a preset number of second reference points after the target moment, and these reference points have corresponding predicted points on the predicted path. Furthermore, constructing the LQR optimization model using these predicted and reference points is equivalent to adding a preview mechanism to the LQR algorithm. When determining the target control parameters, both the current path tracking error and the path tracking error for a future distance are considered. This allows for obtaining more optimal control parameters, avoiding sharp turns, and ensuring vehicle driving stability.

[0076] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following detailed description, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings:

[0078] FIG1 is a flow chart of a vehicle control method according to an embodiment of the present disclosure;

[0079] FIG2 is an exemplary schematic diagram of a reference driving path and a predicted path in the vehicle control method provided by the present disclosure;

[0080] FIG3 is a block diagram of a vehicle control device according to an embodiment of the present disclosure;

[0081] Fig. 4 is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0082] The following describes the specific embodiments of the present disclosure in detail with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to illustrate and explain the present disclosure and are not intended to limit the present disclosure.

[0083] FIG1 is a flow chart of a vehicle control method according to an embodiment of the present disclosure. As shown in FIG1 , the method provided by the present disclosure may include steps 11 to 15.

[0084] In step 11, the vehicle's posture information at the target time and the posture information corresponding to each of the multiple prediction points are determined as the first posture information.

[0085] For example, the target time may be the current time, and accordingly, the first position information is the actual position information of the vehicle at the current time. For another example, the target time may be a time after the current time.

[0086] Prediction points can be determined from the vehicle's predicted path. The vehicle's predicted path typically includes multiple prediction points, generated by predicting the vehicle's path for a series of moments after a target time. Therefore, the first pose information includes the vehicle's pose information at the target time and a series of moments after the target time.

[0087] Optionally, the pose information may include coordinates and heading angles. For example, in a two-dimensional rectangular coordinate system constructed by an x-axis and a y-axis, the coordinates in the pose information include an x-axis coordinate and a y-axis coordinate.

[0088] In step 12, a plurality of reference points are determined on a reference driving path corresponding to the vehicle, and the posture information corresponding to each reference point is determined as the second posture information.

[0089] The reference driving path may be directly obtained from other modules (eg, an upstream path planning module, etc.). For example, the reference driving path may be represented by a curve equation.

[0090] The multiple reference points include a first reference point corresponding to the target time and a preset number of second reference points after the target time. In practice, the second reference points are several future reference points, which in this disclosure are equivalent to preview points. Furthermore, each reference point has a corresponding prediction point on the predicted path.

[0091] In one possible implementation, in step 12, determining multiple reference points on the reference driving path of the vehicle may include the following steps:

[0092] Determine the projection point of the vehicle's coordinates at the target time on the reference driving path as the first reference point;

[0093] Taking the location of the first reference point as the starting location and the preset distance as the spacing between adjacent reference points, a preset number of second reference points are determined one by one on the reference driving path.

[0094] Among them, the preset distance is used as the spacing between adjacent reference points. These spacings can be equal to each other, or different from each other, or partially the same and partially different. This disclosure does not strictly limit this.

[0095] As described above, the position information may include coordinates, and thus, the position information of the vehicle at the target time includes the coordinates of the vehicle at the target time.

[0096] By determining the projection point of the vehicle's coordinates at the target time onto the reference driving path, a reference point corresponding to the target time on the reference driving path, i.e., a first reference point, can be located. The projection point can be determined using a conventional method for determining the projection point of a point onto a curve, i.e., by making the line connecting the point and the projection point perpendicular to the tangent line of the projection point on the curve.

[0097] After determining the first reference point, other reference points (i.e., second reference points) on the reference driving path can be determined one by one, starting from the first reference point, with a preset distance between adjacent reference points. For example, the reference driving path can be shown as M0 in FIG2 , where M1 is the first reference point. Then, M2 (the second reference point) can be determined on the reference driving path at a preset distance interval from M1. Then, M2 (the second reference point) can be determined on the reference driving path at a preset distance interval from M1. The number of second reference points continues until the number of second reference points reaches the preset number.

[0098] After determining the above-mentioned multiple reference points, the posture information of each reference point can be determined as the second posture information. As mentioned above, the posture information can include coordinates and heading angles. At the same time, the reference driving path can correspond to a path equation, for example, the above-mentioned curve equation. Based on this, determining the posture information corresponding to each reference point can include the following steps:

[0099] According to the reference driving path, determine the coordinates corresponding to each reference point;

[0100] For each reference point, the heading angle corresponding to the reference point is determined based on the coordinates and path equation corresponding to the reference point.

[0101] After determining the reference points on the reference driving path in the above manner, the coordinates of each reference point can be directly obtained. Then, the heading angle of each reference point is determined separately.

[0102] For example, the heading angle θ corresponding to the i-th reference point Ri can be determined by the following formula: Ri :

[0103] θ Ri =arctan(f′(x Ri ))

[0104] Among them, 0≤i≤preset number, R0 is the first reference point, f(x Ri ) is the path equation, f′(x Ri ) is the derivative of the path equation.

[0105] Returning to FIG1 , in step 13 , an LQR optimization model is constructed based on the first pose information and the second pose information.

[0106] The LQR optimization model may include a target cost function, and the purpose of the present disclosure is to determine the control parameters that can minimize the target cost function. Typically, the target cost function is accompanied by constraints. Therefore, the purpose of the present disclosure is actually to determine the control parameters that can minimize the target cost function while satisfying the constraints (in this disclosure, the target constraints).

[0107] For example, the control parameter may be the steering curvature.

[0108] Optionally, the target cost function can be determined by:

[0109] Get the pose cost coefficient and control cost coefficient;

[0110] Generate a first cost function according to the first pose information, the second pose information and the pose cost coefficient;

[0111] Generate a second cost function according to the preset control parameters and control cost coefficients;

[0112] The sum of the first cost function and the second cost function is determined as a target cost function.

[0113] The pose cost coefficient and control cost coefficient can be set according to the needs of the actual scenario. If the pose information includes coordinates and heading angles, the pose cost coefficient can include a position cost coefficient and a heading cost coefficient. The sum of the position cost coefficient and the heading cost coefficient maintains a stable value. That is, if the position cost coefficient is increased, the heading cost coefficient is correspondingly reduced.

[0114] By setting the cost coefficient, you can balance vehicle control accuracy and stability. For example, if you set a higher heading cost coefficient, the vehicle's heading will be more accurate during control, but the position deviation will be slightly larger. For another example, if you set a higher control cost coefficient, the vehicle's steering control will be smoother during control, but the accuracy will be reduced.

[0115] Optionally, the target cost function F can be generated according to the following formula:

[0116] in, is the first cost function, is the second cost function; X i is the pose information predicted for the i-th prediction point, X Ri is the second pose information of the i-th reference point, X R0is the second pose information of the first reference point, N is the preset number, Q is the diagonal matrix generated according to the pose cost coefficient; r is the control cost coefficient, κ i is the control parameter of the i-th prediction point.

[0117] As described above, a vehicle may correspond to a predicted path including prediction points. The predicted path may be shown as E0 in FIG2 , where each reference point has a corresponding prediction point on the predicted path, e.g., reference point M2 corresponds to prediction point E2, and reference point M4 corresponds to prediction point E4.

[0118] For example, Q can be generated according to the following formula:

[0119] Q = diag[q xy ,q xy ,q θ ]

[0120] Among them, q xy is the position cost coefficient, q θ is the heading cost coefficient.

[0121] Optionally, the LQR optimization model may further include a target constraint condition, which may include a first constraint condition for constraining the generation method of the predicted path and a second constraint condition for constraining X0 to be the first pose information.

[0122] Optionally, the first constraint condition can be generated in the following way:

[0123] Construct a kinematic model with posture information as state quantity and preset control parameters as control quantity;

[0124] Discretizing the kinematic model according to a preset distance to obtain a system equation for generating position information of a second prediction point according to the position information of the first prediction point, where the second prediction point is a next prediction point separated from the first prediction point by a preset distance;

[0125] The system equations are linearized to obtain the processed objective equation as the first constraint condition.

[0126] For example, the kinematic model can be constructed as follows:

[0127] Among them, x and y are coordinates, θ is the heading angle, and κ is the preset control parameter.

[0128] After constructing the kinematic model, the kinematic model is discretized according to a preset distance to obtain a system equation. Optionally, the discretization process can be derived using a trapezoidal integral formula.

[0129] For example, the system equation after discretization can be:

[0130] Among them, X i is the pose information predicted for the i-th prediction point, Δs is the preset distance, θ i is the heading angle corresponding to the i-th predicted point.

[0131] In this disclosure, the use of a kinematic model to derive the LQR system equations improves lateral control accuracy and reduces tracking error relative to the reference driving path. Furthermore, compared to parameters such as the lateral stiffness of the front and rear tires, vehicle mass, and yaw moment of inertia about the center of mass required for dynamic model construction, which are difficult to accurately obtain and easily affected by operating conditions, the parameters required by the kinematic model are independent of operating conditions. This makes it easier to maintain uniformity in subsequent control, improving control accuracy. Furthermore, the kinematic model can also describe large steering conditions, extending its applicability.

[0132] At the same time, the system equations are discretized using a preset distance, so that the system equations obtained after discretization are independent of the vehicle speed, which is conducive to achieving control decoupling of the vehicle's lateral and longitudinal directions, and making the path tracking effect in the low-speed range (e.g., 0-30 km / h) unaffected by the vehicle speed. Among them, the discretization using the trapezoidal integral formula is also conducive to increasing control accuracy.

[0133] After obtaining the above system equations, linearization can be performed through Taylor expansion. For example, the target equation after Taylor expansion can be:

[0134] X i+1 =A i X i +B i κ i +C i

[0135] Among them, A i 、B i 、C i Determined by:

[0136] Among them, θ Ri is the heading angle corresponding to the i-th reference point.

[0137] In step 14, the target control parameters at the next target moment are determined according to the LQR optimization model.

[0138] In a possible implementation, step 14 may include the following steps:

[0139] Determine the feedback gain matrix that can minimize the target cost function while satisfying the target constraints;

[0140] According to the first pose information and the feedback gain matrix, the target control parameters at the next target moment are determined.

[0141] Optionally, a dynamic programming solution may be used to solve the target cost function to obtain a feedback gain matrix.

[0142] And, the target control parameter κ can be determined according to the following formula:

[0143] κ=-K i X i -L i

[0144] Among them, K i , L i is the feedback gain matrix, and:

[0145] q i =-QX Ri

[0146] Among them, P N =Q,F N =q N .

[0147] In step 15, at a moment next to the target moment, the vehicle is controlled according to the target control parameter.

[0148] In a possible implementation, if the target control parameter is the steering curvature, step 15 may include the following steps:

[0149] determining a front wheel steering angle corresponding to the steering curvature;

[0150] The vehicle is controlled according to the front wheel steering angle.

[0151] For example, the front wheel steering angle δ may be determined according to the following formula:

[0152] δ=arctan(κL)

[0153] Where κ is the turning curvature and L is the wheelbase of the vehicle.

[0154] Through the above technical solution, the vehicle's posture information at the target moment and the posture information corresponding to multiple predicted points on the predicted path are determined as the first posture information. Multiple reference points and their respective second posture information are determined on the reference driving path corresponding to the vehicle. Based on the first and second posture information, an LQR optimization model is constructed to determine target control parameters for the moment immediately following the target moment, and the vehicle is then controlled according to the target control parameters at the moment immediately following the target moment. The reference points determined on the reference driving path include both the first reference point corresponding to the target moment and a preset number of second reference points after the target moment, and these reference points have corresponding predicted points on the predicted path. Furthermore, constructing the LQR optimization model using these predicted and reference points is equivalent to adding a preview mechanism to the LQR algorithm. When determining the target control parameters, both the current path tracking error and the path tracking error for a future distance are considered. This allows for obtaining more optimal control parameters, avoiding sharp turns, and ensuring vehicle driving stability.

[0155] FIG3 is a block diagram of a vehicle control device according to an embodiment of the present disclosure. As shown in FIG3 , the device 30 includes:

[0156] A first determination module 31 is configured to determine the vehicle's posture information at a target time and posture information corresponding to a plurality of prediction points as first posture information, wherein the prediction points are determined from a predicted path corresponding to the vehicle;

[0157] a second determining module 32, configured to determine a plurality of reference points on a reference driving path corresponding to the vehicle, and determine pose information corresponding to each reference point as second pose information, wherein the plurality of reference points include a first reference point corresponding to the target moment and a preset number of second reference points after the target moment, and each reference point has a corresponding prediction point on the predicted path;

[0158] A construction module 33 is used to construct an LQR optimization model according to the first posture information and the second posture information;

[0159] A third determining module 34 is configured to determine a target control parameter at a next moment after the target moment according to the LQR optimization model;

[0160] The control module 35 is configured to control the vehicle according to the target control parameter at a moment next to the target moment.

[0161] Optionally, the position information of the vehicle at the target time includes the coordinates of the vehicle at the target time;

[0162] The second determining module 32 includes:

[0163] A first determining submodule is configured to determine a projection point of the coordinates of the vehicle at the target time on the reference driving path as the first reference point;

[0164] The second determining submodule is configured to determine the preset number of second reference points one by one on the reference driving path, with the location of the first reference point as a starting location and a preset distance as a spacing between adjacent reference points.

[0165] Optionally, the reference driving path corresponds to a path equation, and the posture information includes coordinates and heading angles;

[0166] The second determining module 32 includes:

[0167] A third determining submodule is configured to determine the coordinates corresponding to each of the reference points according to the reference driving path;

[0168] The fourth determining submodule is configured to determine, for each reference point, a heading angle corresponding to the reference point according to the coordinates corresponding to the reference point and the path equation.

[0169] Optionally, the heading angle θ corresponding to the i-th reference point Ri is determined by the following formula: Ri :

[0170] θ Ri =arctan(f′(x Ri ))

[0171] Wherein, 0≤i≤preset number, R0 is the first reference point, f(x Ri ) is the path equation, f′(x Ri ) is the derivative of the path equation.

[0172] Optionally, the LQR optimization model includes an objective cost function;

[0173] The target cost function is determined by:

[0174] Get the pose cost coefficient and control cost coefficient;

[0175] Generate a first cost function according to the first pose information, the second pose information and the pose cost coefficient;

[0176] generating a second cost function according to the preset control parameters and the control cost coefficient;

[0177] The sum of the first cost function and the second cost function is determined as the target cost function.

[0178] Optionally, the target cost function F is generated according to the following formula:

[0179] in, is the first cost function, is the second cost function; X i is the pose information predicted for the i-th prediction point, X Ri is the second pose information of the i-th reference point, X R0 is the second pose information of the first reference point, N is the preset number, Q is the diagonal matrix generated according to the pose cost coefficient; r is the control cost coefficient, κ i is the control parameter of the i-th prediction point.

[0180] Optionally, the posture information includes coordinates and heading angles, and the posture cost coefficient includes a position cost coefficient and a heading cost coefficient;

[0181] Q is generated according to the following formula:

[0182] Q = diag[q xy ,q xy ,q θ ]

[0183] Among them, q xy is the position cost coefficient, q θ is the heading cost coefficient.

[0184] Optionally, the LQR optimization model further includes a target constraint condition, wherein the target constraint condition includes a first constraint condition for constraining a generation method of the predicted path and a second constraint condition for constraining X0 to be the first posture information.

[0185] Optionally, the posture information includes coordinates and heading angles;

[0186] The first constraint condition is generated in the following way:

[0187] Construct a kinematic model with posture information as state quantity and preset control parameters as control quantity;

[0188] Discretizing the kinematic model according to a preset distance to obtain a system equation for generating position information of a second prediction point according to the position information of the first prediction point, where the second prediction point is a next prediction point separated from the first prediction point by a preset distance;

[0189] The system equation is linearized to obtain a processed target equation as the first constraint condition.

[0190] Optionally, the system equation is:

[0191] Among them, X i is the predicted pose information for the i-th prediction point, Δs is the preset distance, θ i is the heading angle corresponding to the i-th predicted point.

[0192] Optionally, the linearization process is Taylor expansion;

[0193] The objective equation is:

[0194] X i+1 =A i X i +B i κ i +C i

[0195] Among them, A i 、B i 、C i Determined by:

[0196] Among them, θ Ri is the heading angle corresponding to the i-th reference point.

[0197] Optionally, the third determining module 34 includes:

[0198] A fifth determining submodule is configured to determine a feedback gain matrix that enables the target cost function to obtain a minimum value while satisfying the target constraint condition;

[0199] The sixth determination submodule is used to determine the target control parameter at the next moment of the target moment based on the first posture information and the feedback gain matrix.

[0200] Optionally, the fifth determination submodule is used to solve the target cost function using a dynamic programming solution to obtain the feedback gain matrix.

[0201] Optionally, the sixth determination submodule is configured to determine the target control parameter κ according to the following formula:

[0202] κ=-K i X i -L i

[0203] Among them, K i , L i is the feedback gain matrix, and:

[0204] q i =-QX Ri

[0205] Among them, P N =Q,F N =q N .

[0206] Optionally, the target control parameter is steering curvature;

[0207] The control module 35 includes:

[0208] a seventh determining submodule, configured to determine a front wheel steering angle corresponding to the steering curvature;

[0209] A control submodule is used to control the vehicle according to the front wheel steering angle.

[0210] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0211] The present disclosure also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps of the vehicle control method provided in any embodiment of the present disclosure are implemented.

[0212] The present disclosure also provides an electronic device, comprising:

[0213] a memory having a computer program stored thereon;

[0214] A processor is used to execute the computer program in the memory to implement the steps of the vehicle control method provided in any embodiment of the present disclosure.

[0215] The present disclosure also provides a vehicle, comprising the electronic device provided by any embodiment of the present disclosure.

[0216] FIG4 is a block diagram of an electronic device 700 according to an exemplary embodiment. As shown in FIG4 , the electronic device 700 may include a processor 701 and a memory 702. The electronic device 700 may also include one or more of a multimedia component 703, an input / output (I / O) interface 704, and a communication component 705.

[0217] The processor 701 is used to control the overall operation of the electronic device 700 to complete all or part of the steps in the above-mentioned vehicle control method. The memory 702 is used to store various types of data to support the operation of the electronic device 700. Such data may include, for example, instructions for any application or method operating on the electronic device 700, as well as application-related data, such as contact information, sent and received messages, pictures, audio, video, etc. The memory 702 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The multimedia component 703 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 702 or transmitted via the communication component 705. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 704 provides an interface between the processor 701 and other interface modules. The aforementioned other interface modules may be a keyboard, a mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 705 is used for wired or wireless communication between the electronic device 700 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, NB-IOT, eMTC, or other 5G, etc., or a combination of one or more thereof, is not limited here. Therefore, the corresponding communication component 705 may include: a Wi-Fi module, a Bluetooth module, an NFC module, etc.

[0218] In an exemplary embodiment, the electronic device 700 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the above-mentioned vehicle control method.

[0219] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided. When executed by a processor, the program instructions implement the steps of the vehicle control method described above. For example, the computer-readable storage medium may be the aforementioned memory 702 including the program instructions. The program instructions may be executed by the processor 701 of the electronic device 700 to perform the vehicle control method described above.

[0220] In another exemplary embodiment, a computer program product is also provided. The computer program product includes a computer program executable by a programmable device, and has a code portion for executing the above-mentioned vehicle control method when executed by the programmable device.

[0221] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings. However, the present disclosure is not limited to the specific details of the above embodiments. Within the technical concept of the present disclosure, various simple modifications can be made to the technical solutions of the present disclosure, and these simple modifications all fall within the scope of protection of the present disclosure.

[0222] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.

[0223] In addition, the various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.

Claims

1. A vehicle control method, characterized in that: The method comprises: Determine the posture information of the vehicle at the target time and the posture information corresponding to each of the plurality of prediction points as the first posture information, wherein the prediction points are determined from the prediction path corresponding to the vehicle; Determine a plurality of reference points on a reference driving path corresponding to the vehicle, and determine the posture information corresponding to each reference point as the second posture information, wherein the plurality of reference points include a first reference point corresponding to the target moment and a preset number of second reference points after the target moment, and each of the reference points has a corresponding prediction point on the prediction path; Constructing an LQR optimization model according to the first posture information and the second posture information; Determining the target control parameter at the next moment of the target moment according to the LQR optimization model; At a moment next to the target moment, the vehicle is controlled according to the target control parameter.

2. The method according to claim 1, characterized in that The position information of the vehicle at the target time includes the coordinates of the vehicle at the target time; Determining a plurality of reference points on the reference driving path of the vehicle includes: Determine a projection point of the coordinates of the vehicle at the target time on the reference driving path as the first reference point; Taking the position of the first reference point as the starting position and the preset distance as the spacing between adjacent reference points, the preset number of second reference points are determined one by one on the reference driving path.

3. The method according to claim 1 or 2, characterized in that: The reference driving path corresponds to a path equation, and the posture information includes coordinates and heading angles; The determining of the posture information corresponding to each reference point includes: Determine the coordinates corresponding to each of the reference points according to the reference driving path; For each of the reference points, the heading angle corresponding to the reference point is determined according to the coordinates corresponding to the reference point and the path equation.

4. The method according to claim 3, characterized in that Determine the heading angle θ corresponding to the i-th reference point Ri by the following formula Ri : θ Ri =arctan(f′(x Ri )) Wherein, 0≤i≤preset number, R0 is the first reference point, f(x Ri ) is the path equation, f′(x Ri ) is the derivative of the path equation.

5. The method according to any one of claims 1 to 4, characterized in that The LQR optimization model includes an objective cost function; The target cost function is determined by: Get the pose cost coefficient and control cost coefficient; Generate a first cost function according to the first pose information, the second pose information and the pose cost coefficient; Generate a second cost function according to the preset control parameters and the control cost coefficient; The sum of the first cost function and the second cost function is determined as the target cost function.

6. The method according to claim 5, characterized in that The target cost function F is generated according to the following formula: in, is the first cost function, is the second cost function; i is the predicted pose information for the i-th prediction point, X Ri is the second pose information of the i-th reference point, X R0 is the second posture information of the first reference point, N is the preset number, Q is the diagonal matrix generated according to the posture cost coefficient; r is the control cost coefficient, κ i is the control parameter of the i-th prediction point.

7. The method according to claim 6, characterized in that The posture information includes coordinates and heading angles, and the posture cost coefficient includes a position cost coefficient and a heading cost coefficient; Q is generated according to the following formula: Q=diag[q xy ,q xy ,q θ ] Among them, q xy is the position cost coefficient, q θ is the heading cost coefficient.

8. The method according to claim 6 or 7, characterized in that: The LQR optimization model also includes a target constraint condition, which includes a first constraint condition for constraining the generation method of the predicted path and a second constraint condition for constraining X0 to be the first posture information.

9. The method according to claim 8, characterized in that The posture information includes coordinates and heading angles; The first constraint condition is generated in the following way: Construct a kinematic model with posture information as state quantity and preset control parameters as control quantity; According to a preset distance, the kinematic model is discretized to obtain a system equation for generating the posture information of a second prediction point according to the posture information of the first prediction point, wherein the second prediction point is a next prediction point separated from the first prediction point by a preset distance; The system equation is linearized to obtain a processed target equation as the first constraint condition.

10. The method according to claim 9, characterized in that The system equation is: Among them, X i is the predicted pose information for the i-th prediction point, Δs is the preset distance, θ i is the heading angle corresponding to the i-th predicted point.

11. The method according to claim 10, characterized in that The linearization process is Taylor expansion; The objective equation is: X i+1 =A i X i +B i k i +C i Among them, A i , B i , C i Determined by: Among them, θ Ri is the heading angle corresponding to the i-th reference point.

12. The method according to claim 11, characterized in that Determining the target control parameter at the next moment of the target moment according to the LQR optimization model includes: Determine a feedback gain matrix that can minimize the target cost function while satisfying the target constraint condition; Determine a target control parameter at a next moment of the target moment according to the first posture information and the feedback gain matrix.

13. The method according to claim 12, characterized in that The determining of a feedback gain matrix that enables the target cost function to obtain a minimum value when the target constraint condition is satisfied includes: The objective cost function is solved by using a dynamic programming solution to obtain the feedback gain matrix.

14. The method according to claim 12 or 13, characterized in that The step of determining the target control parameter at the next moment of the target moment according to the first posture information and the feedback gain matrix includes: The target control parameter κ is determined according to the following formula: κ=-K i X i -L i Among them, K i , L i is the feedback gain matrix, and: q i =-QX Ri Among them, P N =Q, F N =q N .

15. The method according to any one of claims 1 to 14, characterized in that The target control parameter is the steering curvature; The controlling the vehicle according to the target control parameter includes: determining a front wheel steering angle corresponding to the steering curvature; The vehicle is controlled according to the front wheel steering angle.

16. A vehicle control device, characterized in that: The device comprises: A first determination module is used to determine the posture information of the vehicle at the target time and the posture information corresponding to each of the plurality of prediction points as the first posture information, wherein the prediction points are determined from the prediction path corresponding to the vehicle; A second determination module is used to determine a plurality of reference points on a reference driving path corresponding to the vehicle, and determine the posture information corresponding to each reference point as the second posture information, wherein the plurality of reference points include a first reference point corresponding to the target moment and a preset number of second reference points after the target moment, and each of the reference points has a corresponding prediction point on the predicted path; A construction module, configured to construct an LQR optimization model according to the first posture information and the second posture information; A third determination module, used to determine the target control parameter at the next moment of the target moment according to the LQR optimization model; A control module is used to control the vehicle according to the target control parameter at a moment next to the target moment.

17. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the vehicle control method described in any one of claims 1 to 15 are implemented.

18. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the vehicle control method according to any one of claims 1 to 15.

19. A vehicle, characterized in that: An electronic device comprising the electronic device described in claim 18.

Images