An efficient lateral control method, system, and vehicle for autonomous vehicles

Abstract

This invention discloses an efficient lateral control method, system, and vehicle for autonomous vehicles, relating to the field of vehicle lateral control technology. The method includes: during autonomous driving, acquiring the vehicle's current real-time position information, determining reference speed, reference heading angle, and reference curvature based on a preset reference trajectory, and calculating the current state variables; determining the discrete intervals of the reference speed, reference heading angle, and reference curvature within their respective operating ranges based on an offline LQR control gain table, determining the control gain corresponding to the left and right endpoints of the discrete interval for each reference value, and thus determining the control gain interval; calculating the control gain through linear interpolation based on the position of each reference value within its discrete interval; calculating the control law based on the control gain corresponding to each reference value and the current state variables, and performing lateral control of the vehicle according to the control law. This invention ensures computational accuracy and improves computational speed, achieving superior lateral control performance.

Description

Technical Field

[0001] This invention relates to the field of vehicle lateral control technology, and in particular to an efficient lateral control method, system and vehicle for autonomous vehicles. Background Technology

[0002] Lateral control systems, as a key function of autonomous vehicles, can effectively improve driving comfort and safety. The Linear Quadratic Regulator (LQR) control algorithm is a widely used method in the control design and analysis of autonomous vehicles, ensuring accurate trajectory tracking by calculating the steering wheel angle using the LQR controller. However, in practical applications, the calculation of the LQR control algorithm is performed through Riccati iteration. The number of iterations is closely related to system stability. A higher number of iterations allows the vehicle to achieve tracking quickly, but also consumes excessive computation time, resulting in a low frequency of control commands being sent to the chassis for execution, ultimately leading to poor lateral control performance. Conversely, too few iterations require a longer time to achieve trajectory tracking within the allowable error range, resulting in lower tracking accuracy. Therefore, existing LQR-based vehicle lateral control methods, with their online iterative calculation of the LQR control law, struggle to achieve satisfactory lateral control performance. Summary of the Invention

[0003] To address the shortcomings of the existing technology, this invention provides an efficient lateral control method, system, and vehicle for autonomous vehicles. By pre-constructing an offline LQR control gain table, the LQR control gain table can be directly invoked during the actual driving process of the autonomous vehicle. The LQR control law can be quickly obtained by looking up the table, and then the steering wheel angle can be calculated for lateral control. This method can effectively improve the calculation speed while ensuring calculation accuracy, ultimately achieving a better lateral control effect.

[0004] In a first aspect, the present invention provides an efficient lateral control method for autonomous vehicles.

[0005] An efficient lateral control method for autonomous vehicles includes:

[0006] During the autonomous driving process, the vehicle's current real-time location information is obtained, and combined with the preset reference trajectory, the reference speed, reference heading angle and reference curvature are determined to calculate the current state variables;

[0007] Based on the offline LQR control gain table, the discrete intervals of the reference speed, reference heading angle and reference curvature within their respective working ranges are determined, and the control gain corresponding to the left and right endpoints of the discrete interval of each reference value is determined, thereby determining the control gain interval.

[0008] Based on the location of the reference speed, reference heading angle, and reference curvature in the discrete interval, linear interpolation is performed in the control gain interval corresponding to each reference value to obtain the final control gain.

[0009] The control law is calculated based on the control gain corresponding to the reference speed, reference heading angle, and reference curvature, as well as the current state variable. The vehicle is then laterally controlled according to the control law.

[0010] Secondly, the present invention provides an efficient lateral control system for autonomous vehicles.

[0011] An efficient lateral control system for autonomous vehicles includes:

[0012] The data acquisition module is used to acquire the vehicle's current real-time location information during the autonomous driving process, and combine it with a preset reference trajectory to determine the reference speed, reference heading angle and reference curvature, and calculate the current state variables.

[0013] The data analysis module is used to determine the discrete intervals of the reference speed, reference heading angle, and reference curvature within their respective working ranges based on the offline LQR control gain table, and to determine the control gain corresponding to the left and right endpoints of the discrete interval of each reference value, thereby determining the control gain interval; based on the position of the discrete intervals of the reference speed, reference heading angle, and reference curvature, linear interpolation is performed in the control gain interval corresponding to each reference value to obtain the final control gain.

[0014] The lateral control module is used to calculate the control law based on the control gain corresponding to the reference speed, reference heading angle, and reference curvature, as well as the current state variable, and to perform lateral control on the vehicle according to the control law.

[0015] Thirdly, the present invention also provides an electronic device, including a memory and a processor, and computer instructions stored in the memory and running on the processor, wherein the computer instructions, when executed by the processor, perform the steps of the method described in the first aspect.

[0016] Fourthly, the present invention also provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the steps of the method described in the first aspect.

[0017] Fifthly, the present invention also provides a vehicle that employs an efficient lateral control method for autonomous vehicles as described in the first aspect for lateral control, or includes an efficient lateral control system for autonomous vehicles as described in the first aspect.

[0018] The above one or more technical solutions have the following beneficial effects:

[0019] 1. This invention provides an efficient lateral control method, system, and vehicle for autonomous vehicles. By pre-constructing an offline LQR control gain table, the LQR control gain table can be directly called during the actual driving process of the autonomous vehicle. The LQR control law can be quickly obtained by looking up the table, and then the steering wheel angle can be calculated for lateral control. This method can effectively improve the calculation speed while ensuring calculation accuracy, and ultimately achieve a better lateral control effect.

[0020] 2. The core of the efficient lateral control method, system, and vehicle for autonomous vehicles provided by this invention lies in the pre-construction and creation of an offline LQR control gain table. In this invention, a vehicle kinematic model is constructed to establish an LQR control model for lateral tracking control. Based on this LQR control model, the control gain corresponding to the endpoints of each discretized interval of the reference speed, reference heading angle, and reference curvature within the working range is calculated and stored to obtain an offline LQR control gain table. This provides a data foundation for the subsequent efficient lateral control of autonomous vehicles, achieving efficient and rapid lateral control. Attached Figure Description

[0021] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0022] Figure 1 This is a flowchart of the efficient lateral control method for autonomous vehicles according to an embodiment of the present invention;

[0023] Figure 2 This is a flowchart illustrating the construction of an offline LQR control gain table in an embodiment of the present invention. Detailed Implementation

[0024] It should be noted that the following detailed descriptions are exemplary and are intended only to describe specific embodiments and to provide further explanation of the invention, and are not intended to limit the scope of exemplary embodiments of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0025] Example 1

[0026] To address the problems existing in the prior art, this embodiment provides an efficient lateral control method for autonomous vehicles. By pre-constructing an offline LQR control gain table, the LQR control gain table is directly invoked during the actual driving process of the autonomous vehicle, and fast and accurate lateral control of the vehicle is achieved through table lookup.

[0027] The core of the method proposed in this embodiment lies in the creation of an offline LQR control gain table. Specifically, it involves: First, determining the parameters such as reference velocity (ref_v), reference heading angle (ref_heading), and reference curvature (ref_kappa) in the LQR control model, clarifying the working range of these parameters, and discretizing the reference velocity, reference heading angle, and reference curvature into several small intervals within the working range of these parameters. Taking the reference value at the endpoint of each interval and substituting it into the model, the control gain is obtained through Riccati iteration. Second, for each interval endpoint of reference velocity, each interval endpoint of reference heading angle, and each interval endpoint of reference curvature, the corresponding control gain is calculated, stored, and then an offline LQR control gain table is constructed. When using an offline LQR gain table, the left and right endpoints of the discrete interval are determined based on the vehicle's current reference speed, reference heading angle, and reference curvature. The offline LQR control gain table is then used to determine the gain values ​​corresponding to the left and right endpoints of the interval. Linear interpolation is then used to obtain the final corresponding gain value. The control law is determined based on this gain value, and the vehicle is then laterally controlled according to the control law, thereby achieving efficient and accurate lateral vehicle control.

[0028] This embodiment provides an efficient lateral control method for autonomous vehicles, comprising the following steps:

[0029] During the autonomous driving process, the vehicle's current real-time location information is obtained, and combined with the preset reference trajectory, the reference speed, reference heading angle and reference curvature are determined to calculate the current state variables;

[0030] Based on the offline LQR control gain table, the discrete intervals of the reference speed, reference heading angle and reference curvature within their respective working ranges are determined, and the control gain corresponding to the left and right endpoints of the discrete interval of each reference value is determined, thereby determining the control gain interval.

[0031] Based on the location of the reference speed, reference heading angle, and reference curvature in the discrete interval, linear interpolation is performed in the control gain interval corresponding to each reference value to obtain the final control gain.

[0032] The control law is calculated based on the control gain corresponding to the reference speed, reference heading angle, and reference curvature, as well as the current state variable. The vehicle is then laterally controlled according to the control law.

[0033] The following content provides a more detailed description of the efficient lateral control method for autonomous vehicles proposed in this embodiment.

[0034] The core of the efficient lateral control method for autonomous vehicles proposed in this embodiment lies in the pre-construction of an offline LQR control gain table, such as... Figure 2 As shown, the construction of this LQR control gain table includes the following steps:

[0035] Step S1.1: Based on the vehicle kinematics model, establish the LQR control model for vehicle lateral tracking control. The model is as follows:

[0036] e(k+1)=Ae(k)+Bu(k);

[0037] In the above formula, e(k) represents the vehicle's state variables at the current time, i.e., time k. These state variables include the error values ​​of the vehicle's horizontal and vertical coordinates and the error value of the vehicle's heading angle. This error value is the difference between the actual value and the reference value, i.e.: e(k) = (x r (k)-x,y r (k)-y,heading r (k)-heading), e(k+1) represents the vehicle's state variable at the next time step (k+1), u(k) represents the control variable at the current time step (k), and A and B are both system matrices, represented as follows:

[0038]

[0039] The actual position coordinates are represented by (x, y) on both sides, and the actual heading angle, actual curvature, and actual velocity are represented by heading, v, and a, respectively. The reference position coordinates are represented by (x, y) on both sides. r ,y r The reference heading angle, reference curvature, reference velocity, and reference acceleration are respectively represented as heading angle, reference curvature, reference velocity, and reference acceleration. r κ r v r a r δ r =arctan(L*κ r ), The reference heading angle is represented by δ, where L represents the vehicle axle length. r T represents the steering angle of the vehicle's front wheels. s This represents the sampling time when the continuous model is discretized.

[0040] Step S1.2: Obtain the vehicle's actual position information and obtain the vehicle's reference values ​​within the preset reference trajectory. The actual position signal includes actual position coordinates, actual heading angle, actual curvature, actual velocity, and actual acceleration; the reference values ​​include reference position coordinates, reference heading angle, reference curvature, reference velocity, and reference acceleration.

[0041] In practice, the vehicle's autonomous driving lateral control system receives a trajectory from the upper planning layer. This trajectory is a preset reference trajectory, which can be represented by the aforementioned reference values. That is, the trajectory typically includes (x... r ,y r ,heading r ,κ r ,v r ,a r ), (x r ,y r ,heading r ,κ r ,v r ,a r The x, y, heading, v, and a values ​​represent the reference position of the trajectory in the global coordinate system, respectively. Similarly, the current position information of the vehicle, which can be obtained through the vehicle positioning system, is (x, y, heading, v, a).

[0042] Step S1.3: Based on the actual position coordinates, actual heading angle, reference position coordinates, and reference heading angle of the vehicle obtained above, calculate the vehicle's state variables, i.e., according to the following formula:

[0043] e(k)=(x r (k)-x,y r (k)-y,heading r (k)-heading);

[0044] The state variable e(k) of the vehicle is calculated.

[0045] Step S1.4: Based on the preset reference trajectory, determine the working range of reference speed, reference heading angle, and reference curvature, and discretize the reference speed, reference heading angle, and reference curvature within the working range.

[0046] In this embodiment, the reference velocity v r The working range is [0,14], and the unit is m / s; the reference heading angle is... r The working range is [-pi, +pi], in rad; the reference curvature κ r The working range is [-0.125, 0.125], and the unit is 1 / m.

[0047] Step S1.5: After determining the above model, the discrete points of each reference speed, reference heading angle, and reference curvature, as well as the vehicle's state variables, are substituted into the LQR control model. The parameter matrices Q and R in the model are set, and the control gain of the vehicle's position coordinates and heading angle under different reference values ​​is obtained through iterative calculation, thereby constructing an offline LQR control gain table. The parameter matrices Q and R in the model are pre-set manually, and the optimal set of Q and R values ​​is determined through multiple repeated tests as the final values.

[0048] The LQR control gain table described above can be constructed using the following pseudocode:

[0049]

[0050] Finally, the calculated gain files k1.dat, k2.dat, and k3.dat are saved to construct an offline LQR control gain table for later use.

[0051] In this embodiment, during the vehicle's autonomous driving process, an offline LQR control gain table is used for efficient and precise lateral control of the vehicle, such as... Figure 1 As shown, it specifically includes:

[0052] Step S2.1: Utilize the autonomous vehicle's positioning system to obtain the vehicle's current real-time location information. This real-time location information includes real-time position coordinates (including the x and y coordinates), real-time heading angle, real-time velocity, and real-time acceleration (x, y, heading, v, a). Based on this, combine the preset reference trajectory (x...) provided by the planning layer... r ,y r ,heading r ,κ r ,v r ,a r ), determine the reference speed, reference heading angle and reference curvature, and use them to calculate the current state variable e(k).

[0053] Step S2.2: Based on the offline LQR control gain table, determine the reference speed v by looking up the table according to the determined reference speed, reference heading angle, and reference curvature. r Reference heading angle r and reference curvature κ rWithin their respective operating ranges, the discrete intervals are determined, specifically the discrete intervals within the lists ref_v(p), ref_heading(q), and ref_kappa(r). Based on this, the control gain corresponding to the left and right endpoints of the discrete interval for each reference value is determined, and the corresponding control gain values ​​are denoted as [k1_l,k1_r], [k2_l,k2_r], and [k3_l,k3_r], thus defining the control gain interval.

[0054] Step S2.3: Based on the reference speed v r Reference heading angle r Reference curvature κ r The position of the discrete interval is used to perform linear interpolation in the control gain interval corresponding to each reference value to obtain the final control gain, i.e., k1, k2, k3.

[0055] The final control gain obtained through linear interpolation is as follows: First, calculate the reference velocity v. r Reference heading angle r and reference curvature κ r The total distance from these three reference parameters to the left and right endpoints of their respective discrete intervals is denoted as d. l d r Secondly, using the ratio of the total distance between the left and right endpoints as the slope, for each reference parameter, the difference between the gain values ​​corresponding to the left and right endpoints of the discrete interval where the reference parameter is located is multiplied by the slope. The sum of this product and the gain value corresponding to the left endpoint of the discrete interval where the reference parameter is located is used as the control gain corresponding to that reference parameter. For example, the current reference speed v... r Reference heading angle r and reference curvature κ r Let v_x, heading_x, and kappa_x be the reference values. The intervals for v_x, heading_x, and kappa_x are determined to be [v1, v2], heading_x [heading1, heading2], and kappa_x [kappa1, kappa2]. Given that the control gain matrix K = [k1_l, k2_l, k3_l] corresponds to reference speed, reference heading angle, and reference curvature v1, heading1, and kappa1 respectively, and the control gain matrix K = [k2_r, k2_r, k3_r] corresponds to reference speed, reference heading angle, and reference curvature v2, heading2, and kappa2 respectively, the following formula is used for calculation:

[0056] d l=sqrt((v_x-v1)^2+(heading_x-heading1)^2+(kappa_x-kappa1)^2);

[0057] d r =sqrt((v2-v1)^2+(heading2-heading1)^2+(kappa2-kappa1)^2);

[0058] k1_x=k1_l+(k1_r-k1_l)*(d l / d r );

[0059] k2_x=k2_l+(k2_r-k2_l)*(d l / d r );

[0060] k3_x=k3_l+(k3_r-k3_l)*(d l / d r );

[0061] The above calculations yield the control gains k1_x, k2_x, and k3_x corresponding to the reference velocity v_x, reference heading angle heading_x, and reference curvature kappa_x, respectively, thus obtaining k1, k2, and k3.

[0062] Step S2.4: Based on the reference speed v r Reference heading angle r Reference curvature κ r Using the corresponding control gains and the current state variable e(k), the control law u = (k1, k2, k3) is calculated. T *e(k) performs lateral control of the vehicle according to the control law.

[0063] The above method for calling the offline LQR control gain table can be implemented using the following pseudocode:

[0064]

[0065] The control law is quickly calculated using the above method. Based on the control law, the vehicle is laterally controlled, achieving efficient and precise lateral vehicle control.

[0066] Example 2

[0067] This embodiment provides an efficient lateral control system for autonomous vehicles, including:

[0068] The data acquisition module is used to acquire the vehicle's current real-time location information during the autonomous driving process, and combine it with a preset reference trajectory to determine the reference speed, reference heading angle and reference curvature, and calculate the current state variables.

[0069] The data analysis module is used to determine the discrete intervals of the reference speed, reference heading angle, and reference curvature within their respective working ranges based on the offline LQR control gain table, and to determine the control gain corresponding to the left and right endpoints of the discrete interval of each reference value, thereby determining the control gain interval; based on the position of the discrete intervals of the reference speed, reference heading angle, and reference curvature, linear interpolation is performed in the control gain interval corresponding to each reference value to obtain the final control gain.

[0070] The lateral control module is used to calculate the control law based on the control gain corresponding to the reference speed, reference heading angle, and reference curvature, as well as the current state variable, and to perform lateral control on the vehicle according to the control law.

[0071] Example 3

[0072] This embodiment provides an electronic device, including a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When the processor executes the computer instructions, it completes the steps in the efficient lateral control method for an autonomous vehicle as described above.

[0073] Example 4

[0074] This embodiment also provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the steps in the efficient lateral control method for an autonomous vehicle as described above.

[0075] Example 5

[0076] This embodiment also provides a vehicle that uses the above-described efficient lateral control method for autonomous vehicles for lateral control, or includes the above-described efficient lateral control system for autonomous vehicles.

[0077] The steps and methods involved in Embodiments 2 to 5 above correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0078] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0079] The above description is only a preferred embodiment of the present invention. Although the specific implementation of the present invention has been described in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that, based on the technical solution of the present invention, various modifications or variations that can be made by those skilled in the art without creative effort are still within the scope of protection of the present invention.

Claims

1. A method for efficient lateral control of an autonomous vehicle, characterized in that, include: During the autonomous driving process, the vehicle's current real-time location information is obtained, and combined with the preset reference trajectory, the reference speed, reference heading angle and reference curvature are determined to calculate the current state variables; Based on the offline LQR control gain table, the discrete intervals of the reference speed, reference heading angle and reference curvature within their respective working ranges are determined, and the control gain corresponding to the left and right endpoints of the discrete interval of each reference value is determined, thereby determining the control gain interval. Based on the location of the reference speed, reference heading angle, and reference curvature in the discrete interval, linear interpolation is performed in the control gain interval corresponding to each reference value to obtain the final control gain. The control law is calculated based on the control gain corresponding to the reference speed, reference heading angle, and reference curvature, as well as the current state variable. The vehicle is then laterally controlled based on the control law. The construction of the offline LQR control gain table includes: Based on the vehicle kinematics model, an LQR control model for lateral tracking control of the vehicle is established; the LQR control model is as follows: ； In the above formula, Indicates the current time, i.e. k The vehicle's state variables at any given time include the error values ​​of the vehicle's horizontal and vertical coordinates and the error value of the vehicle's heading angle. These error values ​​are the differences between the actual and reference values. , Indicates the next moment, i.e. k The vehicle's state variable at time +1 Indicates the current time, i.e. k The amount of control at any given moment , All are system matrices, and are represented as follows: ， ； Wherein, the x and y coordinates of the actual position are represented as: The actual heading angle, actual curvature, and actual speed are respectively expressed as: The x and y coordinates of the reference position are represented as follows: The reference heading angle, reference curvature, reference velocity, and reference acceleration are respectively expressed as: , , Indicates the reference heading angle. Indicates the vehicle's axle length. Indicates the steering angle of the vehicle's front wheels. This represents the sampling time when the continuous model is discretized; The system acquires the vehicle's actual position information and obtains the vehicle's reference values ​​within a preset reference trajectory. The actual position information includes actual position coordinates, actual heading angle, actual curvature, actual velocity, and actual acceleration. The reference values ​​include reference position coordinates, reference heading angle, reference curvature, reference velocity, and reference acceleration. Calculate the vehicle's state variables based on the vehicle's actual position coordinates, actual heading angle, reference position coordinates, and reference heading angle; Based on a preset reference trajectory, the working range of reference speed, reference heading angle and reference curvature is determined, and the reference speed, reference heading angle and reference curvature within the working range are discretized. The discrete points of each reference speed, reference heading angle, and reference curvature, as well as the vehicle's state variables, are substituted into the LQR control model. The parameter matrices Q and R in the model are set, and the control gain of the vehicle's position coordinates and heading angle under different reference values ​​is obtained through iterative calculation. In this way, an offline LQR control gain table is constructed.

2. The efficient lateral control method for an autonomous vehicle as described in claim 1, characterized in that, The real-time location information includes real-time location coordinates, real-time heading angle, real-time velocity, and real-time acceleration; the location coordinates include the horizontal and vertical coordinates of the location.

3. The efficient lateral control method for an autonomous vehicle as described in claim 1, characterized in that, The vehicle's state variables include the error values ​​of the vehicle's horizontal and vertical coordinates and the error value of the vehicle's heading angle, wherein the error value is the difference between the actual value and the reference value.

4. A high-efficiency lateral control system for autonomous vehicles, characterized in that, include: The data acquisition module is used to acquire the vehicle's current real-time location information during the autonomous driving process, and combine it with a preset reference trajectory to determine the reference speed, reference heading angle and reference curvature, and calculate the current state variables. The data analysis module is used to determine the discrete intervals of the reference speed, reference heading angle, and reference curvature within their respective working ranges based on the offline LQR control gain table, and to determine the control gain corresponding to the left and right endpoints of the discrete interval of each reference value, thereby determining the control gain interval; based on the position of the discrete intervals of the reference speed, reference heading angle, and reference curvature, linear interpolation is performed in the control gain interval corresponding to each reference value to obtain the final control gain. The lateral control module is used to calculate the control law based on the control gain corresponding to the reference speed, reference heading angle, and reference curvature, as well as the current state variable, and to perform lateral control of the vehicle according to the control law. The construction of the offline LQR control gain table includes: Based on the vehicle kinematics model, an LQR control model for lateral tracking control of the vehicle is established; the LQR control model is as follows: ； In the above formula, Indicates the current time, i.e. k The vehicle's state variables at any given time include the error values ​​of the vehicle's horizontal and vertical coordinates and the error value of the vehicle's heading angle. These error values ​​are the differences between the actual and reference values. , Indicates the next moment, i.e. k The vehicle's state variable at time +1 Indicates the current time, i.e. k The amount of control at any given moment , All are system matrices, and are represented as follows: ， ； Wherein, the x and y coordinates of the actual position are represented as: The actual heading angle, actual curvature, and actual speed are respectively expressed as: The x and y coordinates of the reference position are represented as follows: The reference heading angle, reference curvature, reference velocity, and reference acceleration are respectively expressed as: , , Indicates the reference heading angle. Indicates the vehicle's axle length. Indicates the steering angle of the vehicle's front wheels. This represents the sampling time when the continuous model is discretized; The system acquires the vehicle's actual position information and obtains the vehicle's reference values ​​within a preset reference trajectory. The actual position information includes actual position coordinates, actual heading angle, actual curvature, actual velocity, and actual acceleration. The reference values ​​include reference position coordinates, reference heading angle, reference curvature, reference velocity, and reference acceleration. Calculate the vehicle's state variables based on the vehicle's actual position coordinates, actual heading angle, reference position coordinates, and reference heading angle; Based on a preset reference trajectory, the working range of reference speed, reference heading angle and reference curvature is determined, and the reference speed, reference heading angle and reference curvature within the working range are discretized. The discrete points of each reference speed, reference heading angle, and reference curvature, as well as the vehicle's state variables, are substituted into the LQR control model. The parameter matrices Q and R in the model are set, and the control gain of the vehicle's position coordinates and heading angle under different reference values ​​is obtained through iterative calculation. In this way, an offline LQR control gain table is constructed.

5. An electronic device, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, complete the steps of an efficient lateral control method for an autonomous vehicle as described in any one of claims 1-3.

6. A computer-readable storage medium, characterized in that, Used to store computer instructions, which, when executed by a processor, complete the steps of an efficient lateral control method for an autonomous vehicle as described in any one of claims 1-3.

7. A vehicle, characterized in that, Lateral control is performed using an efficient lateral control method for autonomous vehicles as described in any one of claims 1-3, or includes an efficient lateral control system for autonomous vehicles as described in claim 4.

Images