Vehicle lateral motion control method and system

Abstract

The application provides a vehicle lateral motion control method and a vehicle lateral motion control system. The vehicle lateral motion control method comprises: calculating a feedback steering angle based on an LQR algorithm; calculating a feedforward steering angle based on a preview tracking control; and adding the feedback steering angle and the feedforward steering angle to obtain a front wheel steering angle original value for participating in vehicle lateral motion control. In this way, on the one hand, the advantages of simple calculation and less occupation of computing resources of the LQR algorithm and the preview tracking control algorithm can be exerted, and on the other hand, the combination of the calculation results of the two algorithms can effectively improve the control effect. The vehicle lateral motion control system is hardware for running the above control method. Both of them can solve the problem that the existing vehicle lateral motion online control method cannot meet the needs of actual application scenarios.

Description

Technical Field

[0001] This invention relates to the field of vehicle kinematics control technology, and in particular to a method and system for controlling the lateral motion of a vehicle. Background Technology

[0002] Lane centering and lane change by command (or lane change by lever) are comfort assistance functions in Level 2 ADAS driver assistance systems. These two functions control the vehicle's lateral movement based on lane line information perceived by cameras, assisting the driver in steering. In lane centering, the system controls the vehicle to track the lane centerline and travel along the center of the lane. In lane change by command, after the driver activates the function by flicking the steering lever, the system, after determining there is no risk of collision with obstacles in adjacent lanes, controls the vehicle to move laterally to the left or right adjacent lane, completing the lane change and then centering the vehicle within the corresponding lane. Lateral movement control in driver assistance systems generally uses feedback from the camera perception processing module regarding the vehicle's position deviation and steering angle deviation relative to the reference line (or lane centerline), with the front wheel angle or steering wheel angle as the control output. Common control methods have the following shortcomings and challenges:

[0003] (1) Using actual position deviation and orientation angle deviation as feedback, the deviation is adjusted to 0 by PID control. The main drawback is that the PID parameters are difficult to tune, requiring multiple sets of PID parameters to cover the full speed range. Furthermore, the control objectives of position deviation and orientation angle deviation conflict with each other. The process of reducing position deviation will inevitably increase orientation angle deviation, making it difficult to cope with large deviation conditions.

[0004] (2) The pure tracking algorithm based on the kinematic model of a two-wheeled bicycle and the geometric relationship of the pre-aiming point position is not applicable to high-speed conditions and conditions with large lateral deviations due to the relatively simplified kinematic model. It also cannot eliminate steady-state control deviations and has poor lane centering performance.

[0005] (3) Control based on the dynamic model of a two-wheeled bicycle requires high accuracy of vehicle dynamic parameters (lateral stiffness, moment of inertia, etc.). Inaccurate parameters or model mismatch caused by simplification of the system model can lead to problems such as steering and steering wheel control oscillation, affecting the control effect. If the MPC (model predictive control) method is used, the optimization solution process requires a large amount of computation, while the controller has limited computing and storage resources, making it unsuitable for online control.

[0006] (4) In the lane centering application scenario, the center line of the vehicle lane is directly used as the control reference line. In the lane change application scenario, the center line of the target lane is directly used as the reference line. The initial lateral deviation is large, and the above control method is difficult to cope with. If a transition trajectory is used to switch from the center line of the vehicle lane to the center line of the target lane, a path planning algorithm is also required, which increases the complexity of the algorithm.

[0007] In summary, existing online control methods for vehicle lateral movement cannot meet the needs of practical application scenarios. Summary of the Invention

[0008] The purpose of this invention is to provide a vehicle lateral motion control method and a vehicle lateral motion control system to meet the needs of online vehicle lateral motion control application scenarios.

[0009] To address the aforementioned technical problems, the present invention provides a vehicle lateral motion control method, the vehicle lateral motion control method comprising:

[0010] The feedback steering angle is calculated based on the LQR algorithm; wherein, the control deviation in the LQR algorithm is: the lateral deviation of the vehicle relative to the lane centerline, the rate of change of the lateral deviation of the vehicle relative to the lane centerline, the azimuth deviation of the vehicle relative to the lane centerline, and the rate of change of the azimuth deviation of the vehicle relative to the lane centerline.

[0011] The feedforward rotation angle is calculated based on the aiming and tracking control; the feedforward rotation angle is calculated based on kinematic geometry.

[0012] The original value of the front wheel steering angle is obtained by adding the feedback steering angle and the feedforward steering angle.

[0013] And, the actual steering wheel angle is calculated based on the original value of the front wheel steering angle.

[0014] Optionally, the step of calculating the feedback angle based on the LQR control method includes: determining a scaling factor based on the operating state and the vehicle's kinematic parameters, wherein the operating state is one of the following states: lane centering state and lane change command state; and multiplying and adding the feedback gain matrix, the scaling factor, and the control deviation in the LQR algorithm to calculate the feedback angle.

[0015] Optionally, the step of determining the scaling factor based on the operating state and the vehicle's kinematic parameters includes: if the operating state is the commanded lane change state, determining whether the lateral position deviation of the current vehicle is less than a preset deviation threshold based on the kinematic parameters; if it is less, then using the first scaling factor; otherwise, using the second scaling factor.

[0016] If the motion state is the lane centering state, the third scaling factor is used.

[0017] Wherein, the first scaling factor is less than the second scaling factor, and the third scaling factor is less than the second scaling factor.

[0018] Optionally, the step of calculating the feedback steering angle based on the LQR control method includes: determining the operating state based on the turn signal and the enable signal for the lane change or lane centering function.

[0019] Optionally, the step of calculating the feedforward rotation angle based on the pre-aiming tracking control includes:

[0020] The target lane boundary line is obtained from the feature point coordinate sequence or the curve coefficient of the fitted curve based on the camera; wherein, the target lane boundary line includes the boundary lines of the two lanes in a lane: the current lane, the left lane and the right lane.

[0021] Calculate the lane centerline based on the target lane boundary line.

[0022] Select a preview point in front of the center line of the lane.

[0023] Furthermore, the feedforward steering angle is calculated based on the vehicle wheelbase, the relative distance between the aiming point and the vehicle, and the deflection angle between the aiming point and the vehicle.

[0024] Optionally, the step of selecting a preview point in front of the lane centerline includes: selecting the preview point based on the current vehicle speed and the curvature of the lane centerline; the faster the vehicle speed, the farther the preview point; the greater the curvature of the lane centerline, the closer the preview point.

[0025] Optionally, the step of calculating the feedback angle based on the LQR algorithm includes:

[0026] The lateral deviation e1 is obtained by capturing the original lateral deviation of the vehicle relative to the center line of the lane using a camera, and then limiting the original lateral deviation and its slope.

[0027] The vehicle's orientation angle deviation e2 relative to the lane centerline is obtained based on the camera.

[0028] The rate of change of the vehicle's lateral deviation relative to the lane centerline is obtained by multiplying the vehicle speed and e2.

[0029] Additionally, the rate of change of the vehicle's orientation angle deviation relative to the lane centerline is obtained by subtracting the yaw rate from the product of vehicle speed and lane centerline curvature.

[0030] Optionally, the step of calculating the actual steering wheel angle based on the original value of the front wheel steering angle includes:

[0031] The first steering wheel angle control value is calculated using the original value of the front wheel angle and the current steering wheel angle.

[0032] If the amplitude and rate of change of the first steering wheel angle control quantity are both within the preset range, the actual steering wheel angle is calculated directly using the first steering wheel angle control quantity.

[0033] Furthermore, if the amplitude or rate of change of the first steering wheel angle control quantity is not within the preset range, the first steering wheel angle control quantity is limited to obtain the second steering wheel angle control quantity, and the actual steering wheel angle is calculated based on the second steering wheel angle control quantity.

[0034] Optionally, the vehicle lateral motion control method further includes:

[0035] The system determines whether to activate the lateral control function based on the vehicle status, external input commands, and environmental perception.

[0036] If the determination result is "activated", the actual steering wheel angle is output to control the vehicle's movement.

[0037] Furthermore, if the determination result is that the steering wheel is not activated, the actual steering wheel angle will not be output.

[0038] To address the aforementioned technical problems, the present invention also provides a vehicle lateral motion control system, the vehicle lateral motion control system comprising:

[0039] LQR feedback calculation module: used to calculate feedback steering angle based on LQR algorithm; wherein, the control deviation in the LQR algorithm is: lateral deviation of vehicle relative to lane centerline, rate of change of lateral deviation of vehicle relative to lane centerline, azimuth angle deviation of vehicle relative to lane centerline, and rate of change of azimuth angle deviation of vehicle relative to lane centerline.

[0040] Pre-aiming pure tracking feedforward module: calculates feedforward rotation angle based on pre-aiming tracking control; the feedforward rotation angle is calculated based on kinematic geometry.

[0041] Output module: used to obtain the original value of the front wheel steering angle by adding the feedback steering angle and the feedforward steering angle; and used to calculate the actual steering wheel angle based on the original value of the front wheel steering angle.

[0042] Compared with existing technologies, the present invention provides a vehicle lateral motion control method and a vehicle lateral motion control system. The vehicle lateral motion control method includes: calculating a feedback angle based on the LQR algorithm; calculating a feedforward angle based on preview tracking control; and adding the feedback angle and the feedforward angle to obtain the original value of the front wheel steering angle, which is used to participate in the vehicle lateral motion control. This configuration leverages the advantages of the LQR algorithm and the preview tracking control algorithm—simple calculations and low computational resource consumption—while combining their calculation results effectively improves the control performance. The vehicle lateral motion control system is the hardware that runs the above control method. Both methods address the problem that existing online vehicle lateral motion control methods cannot meet the needs of practical application scenarios. Attached Figure Description

[0043] Those skilled in the art will understand that the accompanying drawings are provided to better understand the invention and do not constitute any limitation on the scope of the invention. Wherein:

[0044] Figure 1 This is a schematic flowchart of a vehicle lateral motion control method according to an embodiment of the present invention;

[0045] Figure 2 This is a schematic diagram of the control deviation according to an embodiment of the present invention;

[0046] Figure 3 This is a schematic diagram of a bicycle's dynamics model;

[0047] Figure 4 This is a schematic diagram of the geometric relationship of the pre-aiming feedforward control according to an embodiment of the present invention;

[0048] Figure 5 This is a structural block diagram of a vehicle lateral motion control system according to an embodiment of the present invention;

[0049] Figure 6 This is a real vehicle test result of a vehicle lateral motion control method according to an embodiment of the present invention under a certain working condition;

[0050] Figure 7 This is the first real-vehicle test result of a vehicle lateral motion control method according to an embodiment of the present invention under working condition two;

[0051] Figure 8 This is the second result of a real vehicle test of a vehicle lateral motion control method according to an embodiment of the present invention under working condition two.

[0052] in:

[0053] 1-Left lane boundary line; 2-Right lane boundary line; 3-Lane center line; 4-Feature point; 5-Vehicle; 6-Corresponding point; 7-Center point; 8-Tangent; 9-Parallel line; 10-Preview point.

[0054] 20 - Vehicle lateral motion control system; 21 - Lane change process recognition module; 22 - Target lane line selection module; 23 - LQR feedback deviation and gain coefficient calculation module; 24 - LQR feedback term scaling coefficient selection module; 25 - Pre-aiming pure tracking feedforward module; 26 - LQR feedback calculation module; 27 - Output module. Detailed Implementation

[0055] To make the objectives, advantages, and features of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the drawings are all in a very simplified form and are not drawn to scale, and are only used to facilitate and clarify the explanation of the embodiments of this invention. Furthermore, the structures shown in the drawings are often part of the actual structures. In particular, different figures may emphasize different aspects and may sometimes use different scales.

[0056] As used in this invention, the singular forms “a,” “an,” and “the” include plural objects; the term “or” is generally used to mean “and / or”; the term “a number” is generally used to mean “at least one”; and the term “at least two” is generally used to mean “two or more”. Furthermore, the terms “first,” “second,” and “third” are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as “first,” “second,” or “third” may explicitly or implicitly include one or at least two of that feature. “One end” and “the other end,” as well as “proximal end” and “distal end,” generally refer to two corresponding parts, including not only endpoints. The terms “installed,” “connected,” and “joined” should be interpreted broadly, for example, as a fixed connection, a detachable connection, or an integral part; a mechanical connection or an electrical connection; a direct connection or an indirect connection through an intermediate medium; or a connection within two elements or an interaction between two elements. Furthermore, as used in this invention, the phrase "one element is disposed on another element" generally only indicates that there is a connection, coupling, cooperation, or transmission relationship between the two elements, and the connection, coupling, cooperation, or transmission between the two elements can be direct or indirect through an intermediate element. It should not be construed as indicating or implying a spatial positional relationship between the two elements, i.e., one element can be located arbitrarily inside, outside, above, below, or to one side of the other element, unless otherwise explicitly stated. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0057] The core idea of ​​this invention is to provide a vehicle lateral motion control method and a vehicle lateral motion control system to meet the needs of online vehicle lateral motion control application scenarios.

[0058] The following description refers to the accompanying drawings.

[0059] Please refer to Figure 1 The present invention provides a method for controlling the lateral movement of a vehicle, the method comprising:

[0060] S10 calculates the feedback angle based on the LQR algorithm.

[0061] S20, calculate the feedforward rotation angle based on the pre-aiming tracking control; the feedforward rotation angle is calculated based on kinematic geometry.

[0062] S30, the feedback angle and the feedforward angle are added together to obtain the original value of the front wheel angle.

[0063] And, S40, calculate the actual steering wheel angle based on the original value of the front wheel steering angle.

[0064] In step S10, the control deviation in the LQR algorithm is: the lateral deviation e1 of the vehicle relative to the lane centerline, and the rate of change of the lateral deviation of the vehicle relative to the lane centerline. The vehicle's azimuth deviation relative to the lane centerline e2 and the rate of change of the vehicle's azimuth deviation relative to the lane centerline

[0065] Please refer to Figure 2 In the diagram, XY represents the ground coordinate system, and Vx-Vy represents the vehicle body coordinate system. Figure 2 The diagram shows the left lane boundary line 1 and the right lane boundary line 2, and the lane center line 3 calculated based on these two lines. The lane center line 3 is an imaginary line. In one embodiment, the left lane boundary line 1 and the right lane boundary line 2 are characterized by several feature points 4, or they can be characterized by curve fitting. The vehicle 5 has a center point 7, and there is a point on the lane center line 3 that corresponds to the center point 7 (the correspondence can be the intersection of the center point 7 along the y-axis of the vehicle coordinate system and the lane center line 3, or other reasonable correspondence methods), which can be called the corresponding point 6. The difference between the corresponding point 6 and the center point 7 is the lateral deviation e1 of the vehicle relative to the lane center line. A tangent 8 to the lane center line 3 can be drawn through the corresponding point 6. Translating the tangent 8 to pass through the center point 7 yields a parallel line 9. The angle between the x-axis of the vehicle coordinate system and the parallel line 9 is the directional angle deviation e2 of the vehicle relative to the lane center line. These are the first derivatives of e1 and e2, respectively. It's understandable that the above only introduces the physical meaning of control deviation; in actual control processes, other methods will be used to measure these physical quantities.

[0066] According to the principle of LQR control, the control quantity output by LQR feedback control is the feedback angle δ. fbk Calculate using the following formula: δ fbk =-K T *err.

[0067] Among them, control deviation Feedback gain matrix K = [k1 k2 k3 k4] T By selecting reasonable values ​​for the Q and R matrices, the Riccati equation of the linear quadratic form is solved offline to obtain K, which is then stored in the controller.

[0068] The above process can be calculated using the standard LQR algorithm. The general process is as follows: Find the minimum value of the J-cost function.

[0069]

[0070] Where J is the cost function optimized by LQR, and LQR aims to minimize this cost. The cumulative cost in the formula does not consider the terminal state; the cost to be optimized is the integral accumulation of the system state deviation and the control quantity (front wheel steering angle). A and B in the system equations are obtained from the vehicle dynamics model and parameters; the weight matrices R and Q are adjustable and can be set and calibrated according to actual conditions.

[0071] Based on the HJB conditions, the riccati equation can be obtained:

[0072]

[0073] And the optimal control law is obtained:

[0074] u * (t)=-Q -1 (t)B T (y)P(t)x(t)=-K(t)x(t)

[0075] When the vehicle speed is constant or approximately constant, K can be considered constant, thus allowing for offline calculation and solution. The table showing the relationship between vehicle speed and gain coefficient can be calculated offline and stored in the controller, and obtained by looking up the table based on the vehicle speed during function execution.

[0076] Matrix A and B can be based on Figure 3 To understand, Figure 3 This is a schematic diagram of the bicycle dynamics model of a vehicle. The symbols in the diagram have unified definitions in the field of vehicle dynamics, so they will not be described in detail here. Key parameters are introduced below: δ f The front wheel steering angle is the angle between the direction of the front wheels and the longitudinal axis of the vehicle; α fand α r These are the slip angles of the front and rear wheels, respectively, which are the angles between the vehicle's speed direction and the wheel's position. β is the center-of-gravity slip angle.

[0077] The system equations are as follows:

[0078]

[0079]

[0080] u(t)=[δ f ]

[0081] In this embodiment, unlike existing technologies, adjustable scaling factors w1 to w4 are further added to the feedback gain matrix K, facilitating flexible adjustment of the proportion of each feedback term according to actual application conditions. The feedback angle is calculated as follows:

[0082]

[0083] W = [w1w2w3w4] is called the scaling factor.

[0084] Specifically, the step of calculating the feedback steering angle based on the LQR control method includes: determining a scaling factor based on the operating state and the vehicle's kinematic parameters, wherein the operating state is one of the following states: lane centering state and lane change command state; and multiplying and adding the feedback gain matrix, the scaling factor, and the control deviation in the LQR algorithm to calculate the feedback steering angle.

[0085] Furthermore, the step of determining the scaling factor based on the operating state and the vehicle's kinematic parameters includes: if the operating state is the commanded lane change state, determining whether the lateral position deviation of the current vehicle is less than a preset deviation threshold based on the kinematic parameters; if it is less, then the first scaling factor is used; otherwise, the second scaling factor is used.

[0086] If the motion state is the lane centering state, the third scaling factor is used.

[0087] Wherein, the first scaling factor is less than the second scaling factor, and the third scaling factor is less than the second scaling factor.

[0088] The word "less than" here should be interpreted broadly. For example, it can mean that the sum is less than, the weighted sum is less than, or the final control effect is less than.

[0089] In a specific embodiment, the dominant feedback term is selected by setting scaling factors w1 to w4 under different deviation levels and vehicle speeds. For the lane centering function, the initial lateral deviation is small, and the control objective is to converge the lateral deviation to zero. The influence of the directional angle is not significant, and using the lateral deviation e1 as the main feedback can achieve a better centering effect. For the lane change function, since the initial lateral deviation e1 is large, the lane line directional angle e2 will increase during the process of controlling the steering to reduce e1. If only the lateral deviation e1 is used as feedback, overshoot is likely to occur. This manifests as the vehicle entering the target lane at a large angle when entering the lane change, but the steering wheel returns to center late. The steering wheel is only turned in the opposite direction after crossing the center line of the target lane, resulting in a relatively abrupt lane change process. At higher speeds, the steering wheel may not have enough time to return to center, and the vehicle may run out of the target lane. In addition to the directional angle deviation e2, the deviation of the directional angle is also reflected in the lateral deviation change rate. In the middle, and the higher the speed, the more the heading angle is... The greater the impact, the more this scheme primarily selects the lateral deviation e1 and its rate of change. The orientation angle deviation e2 serves as feedback during the lane-changing process. This feedback introduces... The steering angle in E2 acts as a "damping" mechanism, allowing the steering wheel to return to center in time during lane changes without overshooting, resulting in a smoother lane change process.

[0090] Therefore, for lane changing, the lateral deviation e1 is used as the criterion. If e1 is lower than the preset movement threshold (0.4m), W takes the value [0.5; 0; 0.1; 0]; if e1 is higher than the preset movement threshold, W takes the value [0.5; 0.5; 0.6; 0]. For lane centering, W takes the value [0.5; 0; 0.1; 0]. Different values ​​of the scaling factor satisfy the functional differences between lane centering and lane changing commands. In other embodiments, the specific values ​​of the preset movement threshold and W may differ from this example.

[0091] In one specific embodiment, the step of calculating the feedback steering angle based on the LQR control method includes: determining the operating state based on turn signal signals and enable signals for lane change or lane centering functions. The driver's steering intention and commands are obtained through the turn signal signals, and the lateral control conditions are distinguished into lane change and lane centering.

[0092] The step of calculating the feedforward rotation angle based on the preview tracking control includes:

[0093] The target lane boundary line is obtained from the feature point coordinate sequence or the curve coefficient of the fitted curve based on the camera; wherein, the target lane boundary line includes the boundary lines of the two lanes in a lane: the current lane, the left lane and the right lane.

[0094] Calculate the lane centerline based on the target lane boundary line.

[0095] Select a preview point in front of the center line of the lane.

[0096] Furthermore, the feedforward steering angle is calculated based on the vehicle wheelbase, the relative distance between the aiming point and the vehicle, and the deflection angle between the aiming point and the vehicle.

[0097] The feature point coordinate sequence can be referenced. Figure 2 To understand this, one approach is to fit the curve using a cubic polynomial, then retain the polynomial coefficients for subsequent calculations.

[0098] The geometric relationship of the aiming feedforward control is as follows: Figure 4 As shown. Figure 4 In this context, R represents the turning radius, α represents the deflection angle between the aiming point 10 and the vehicle, ld is the relative distance between the aiming point and the vehicle, and L is the wheelbase of the vehicle. According to... Figure 4 δ can be calculated fwd =arctan(L / R), R = l d / 2sinα. Where, δ fwd This indicates the feedforward rotation angle.

[0099] Furthermore, the step of selecting a preview point in front of the lane centerline includes: selecting the preview point based on the current vehicle speed and the curvature of the lane centerline; the faster the vehicle speed, the farther the preview point; the greater the curvature of the lane centerline, the closer the preview point. Specific preview point selection details can be set according to actual needs, for example, being directly proportional to vehicle speed and inversely proportional to curvature, or only positively correlated with vehicle speed but not in a linear relationship, etc.

[0100] Specifically, the step of calculating the feedback angle based on the LQR algorithm includes:

[0101] Based on the camera's acquisition of the vehicle's original lateral deviation relative to the lane centerline, when the lateral control switches from lane centering to lane change command, the reference centerline changes from the current lane to the adjacent lane, causing a step change in the original lateral deviation. This can lead to a sudden change in steering wheel angle or an excessively large rate of change in steering angle. By limiting the slope of the lateral deviation change, abrupt changes in the lateral deviation e1 input to the algorithm during the transition process are avoided. Furthermore, when the e1 sensed by the camera exceeds a preset upper limit, the preset upper limit is taken as e1. In other words, the lateral deviation e1 is obtained by limiting the original lateral deviation and its change slope.

[0102] The vehicle's orientation angle deviation e2 relative to the lane centerline is obtained based on the camera.

[0103] The rate of change of the vehicle's lateral deviation relative to the lane centerline is obtained by multiplying the vehicle speed and e2. That is, Where v represents vehicle speed.

[0104] Additionally, the rate of change of the vehicle's orientation angle deviation relative to the lane centerline is obtained by subtracting the yaw rate from the product of vehicle speed and lane centerline curvature. That is, Where k represents the curvature of the lane centerline and ω represents the yaw rate.

[0105] In addition, to ensure steering comfort and safety, the amplitude and rate of change of the steering wheel angle control quantity are limited before the front wheel steering angle is converted into the steering wheel angle output. That is, the step of calculating the actual steering wheel angle based on the original front wheel steering angle value includes:

[0106] The first steering wheel angle control value is calculated using the original value of the front wheel angle and the current steering wheel angle.

[0107] If the amplitude and rate of change of the first steering wheel angle control quantity are both within the preset range, the actual steering wheel angle is calculated directly using the first steering wheel angle control quantity.

[0108] Furthermore, if the amplitude or rate of change of the first steering wheel angle control quantity is not within the preset range, the first steering wheel angle control quantity is limited to obtain the second steering wheel angle control quantity, and the actual steering wheel angle is calculated based on the second steering wheel angle control quantity.

[0109] The actual steering wheel angle must also be determined based on the lateral control enable conditions to participate in the final vehicle motion control. This is primarily determined by the current vehicle status, whether the driver has activated the function, and the environmental perception status, and is determined by an external enable condition judgment module. Lateral control can only be enabled under conditions such as the driver activating the function, the vehicle being under control, and the lane markings being clear and complete. The steering wheel angle is only output after lateral control is enabled; otherwise, the lateral control function is not activated.

[0110] That is, the vehicle lateral motion control method further includes:

[0111] The system determines whether to activate the lateral control function based on the vehicle status, external input commands, and environmental perception.

[0112] If the determination result is "activated", the actual steering wheel angle is output to control the vehicle's movement.

[0113] Furthermore, if the determination result is that the steering wheel is not activated, the actual steering wheel angle will not be output.

[0114] This embodiment also provides a vehicle lateral motion control system 20, the vehicle lateral motion control system 20 including:

[0115] Lane change process recognition module 21: This module acquires turn signal signals and enable signals for lane change or lane centering commands, and determines the vehicle's current operating state based on these signals. Specifically, in the lane change command operation state, the driver's turn signal serves as the start of the lane change process, triggering lateral control for the lane change command. If the vehicle changes lanes to the left, the end time of the lane change process is the moment the vehicle crosses the left lane line, enters the left lane, and completes centering. If the vehicle changes lanes to the right, the end time is the moment the vehicle crosses the right lane line, enters the right lane, and completes centering. Crossing the left lane line is indicated by a sudden change in the lateral distance of the left lane line perceived by the camera from near zero to close to the lane width, while the distance of the right lane line changes from close to the lane width to near zero. Crossing the right lane line is indicated by the opposite trends in the left and right lane lines. Centering within the corresponding lane is indicated by a decrease in the lateral position deviation e1 and the orientation angle deviation e2, both remaining within a set threshold.

[0116] Target lane selection module 22: When the vehicle is in the lane change command state, it selects the corresponding target lane boundary line as the lateral control information input based on the lane change direction (left or right) indicated by the turn signal. For left lane changes, the left lane line and the left-left lane line sensed by the camera are selected as the target lane lines; for right lane changes, the right lane line and the right-right lane line are selected as the target lane lines. If the lane change process has not yet begun or the lane lines have been crossed, the default is lane centering, and the left and right lane lines of the vehicle are selected as the target lane lines. The corresponding lane center line can be calculated based on the selected lane lines on both sides.

[0117] LQR feedback deviation and gain coefficient calculation module 23: used to acquire the control deviation, store the gain coefficient (which can also be understood as a vehicle speed-gain coefficient lookup table) and output it to LQR feedback calculation module 26.

[0118] LQR feedback scaling factor selection module 24: used to determine the scaling factor based on the operating status and vehicle kinematic parameters and output it to LQR feedback calculation module 26.

[0119] Pre-aiming pure tracking feedforward module 25: calculates the feedforward rotation angle based on pre-aiming tracking control; the feedforward rotation angle is calculated based on kinematic geometry.

[0120] LQR Feedback Calculation Module 26: Used to calculate the feedback angle based on the LQR algorithm; the calculation method can be found in the previous text.

[0121] Output module 27: used to obtain the original value of the front wheel angle by adding the feedback angle and the feedforward angle; and used to calculate the actual steering wheel angle based on the original value of the front wheel angle.

[0122] The vehicle lateral motion control system also includes a lateral control enabling module (not shown). This module determines whether to activate the lateral control function based on the vehicle status, external input commands, and environmental perception. Lateral control requires the driver to activate the function, the vehicle to be under control, and lane markings to be clear and complete before it can be enabled. The lateral control enabling signal can be either lateral control deactivated or activated. Lateral control activation includes either lane centering or lane change command activation. If the determination result is activation, the output module 27 is driven to output the actual steering wheel angle to control vehicle movement; conversely, if the determination result is deactivated, the output module 27 is driven not to output the actual steering wheel angle.

[0123] This embodiment was tested on a real vehicle, demonstrating its beneficial effects. The specific test conditions and results are as follows.

[0124] Operating Condition 1: Lane Centering Condition. The vehicle is traveling at a speed of 40 km / h, moving along a road with clearly marked lane lines, and the lane centering function is activated.

[0125] Test results are as follows Figure 6 As shown. Figure 6 In the diagram, the green line represents the lane centering enable signal, which changes from 0 to 1 at time t2 (represented by the purple vertical line), initiating lane centering control. The red line represents the target lane centerline deviation. After time t2, lane centering control begins. The initial fluctuations in lane centerline deviation gradually decrease after centering control. The absolute value of the lane centerline deviation remains consistently around 0.1m, a relatively small value, significantly reduced from the initial deviation, indicating that it has been maintained near the lane centerline.

[0126] Condition 2: Lane Change Test. The vehicle is traveling at a speed of 40 km / h, moving along a road with clear lane markings in the first lane. The lane centering function is activated. The driver signals with the turn signal, and the vehicle changes lanes.

[0127] Test results are as follows Figure 7 and Figure 8 As shown. Among them. Figure 7 This shows the changes in lateral deviation. Figure 8 The diagram illustrates the changes in the orientation angle deviation. At times t1 and t2 (represented by the purple vertical lines in the diagram, the former being t1 and the latter t2), the right turn and left turn signals are activated respectively, and the center line of the right or left lane is chosen as the target.

[0128] Figure 7In the diagram, the green line represents the lane centering enable signal, which remains at 1. The red line represents the target lane centerline deviation, which changes at times t1 and t2. After times t1 and t2, the target lane becomes the adjacent lane, and the lane centerline deviation becomes the adjacent lane centerline deviation. The smaller of the absolute values ​​of this deviation and a fixed threshold is taken as e1, which serves as the controller's state input, and the vehicle begins to steer to reduce the deviation. Once the absolute value of the lane centerline deviation falls below the fixed threshold, it gradually decreases, and the vehicle enters the target lane. After the absolute value of the lane centerline deviation gradually decreases and remains within 0.4m, the vehicle can be considered to be within the target lane.

[0129] Figure 8 In the diagram, the green line represents the lane centering enable signal, which remains at 1; the light blue line represents the target lane centerline angle (in radians). After times t1 and t2, the lane centerline deviation becomes the adjacent lane centerline deviation, the vehicle begins to turn, and the absolute value of the target lane centerline angle increases. After the vehicle enters the target lane, the target lane becomes the current lane, and the lane centering function straightens the vehicle, decreasing the absolute value of the target lane centerline angle. Throughout the lane change process, the lane centerline angle remains within a small range, indicating a relatively smooth lane change.

[0130] from Figures 6-8 As can be seen from the content, this embodiment has achieved better control effect in terms of vehicle lateral motion control.

[0131] In summary, this embodiment provides a vehicle lateral motion control method and a vehicle lateral motion control system. The vehicle lateral motion control method includes: calculating a feedback steering angle based on the LQR algorithm; calculating a feedforward steering angle based on preview tracking control; and adding the feedback steering angle and the feedforward steering angle to obtain the original value of the front wheel steering angle, which is used to participate in the vehicle lateral motion control. This configuration leverages the advantages of the LQR algorithm and the preview tracking control algorithm in terms of computational simplicity and low computational resource consumption, while combining the calculation results of the two can effectively improve the control effect. The vehicle lateral motion control system is the hardware that runs the above control method. Both can solve the problem that existing online vehicle lateral motion control methods cannot meet the needs of practical application scenarios.

[0132] The above description is only a description of preferred embodiments of the present invention and is not intended to limit the scope of the present invention in any way. Any changes or modifications made by those skilled in the art based on the above disclosure shall fall within the protection scope of the present invention.

Claims

1. A method for controlling the lateral motion of a vehicle, characterized in that, The vehicle lateral motion control method includes: The feedback steering angle is calculated based on the LQR algorithm; wherein, the control deviation in the LQR algorithm is: the lateral deviation of the vehicle relative to the lane centerline, the rate of change of the lateral deviation of the vehicle relative to the lane centerline, the directional angle deviation of the vehicle relative to the lane centerline, and the rate of change of the directional angle deviation of the vehicle relative to the lane centerline. The feedforward rotation angle is calculated based on the aiming and tracking control; the feedforward rotation angle is calculated based on kinematic geometry. The original value of the front wheel steering angle is obtained by adding the feedback steering angle and the feedforward steering angle; and... The actual steering wheel angle is calculated based on the original value of the front wheel steering angle. The steps for calculating the feedback rotation angle based on the LQR algorithm include: The scaling factor is determined based on the operating state and the vehicle's kinematic parameters, wherein the operating state is one of the following states: lane centering state and instructed lane change state; and, The feedback angle is calculated by multiplying and summing the feedback gain matrix, the scaling factor, and the control deviation in the LQR algorithm. The step of determining the scaling factor based on the operating state and the vehicle's kinematic parameters includes: If the operating state is the lane change command state, determine whether the lateral position deviation of the current vehicle is less than a preset deviation threshold based on the kinematic parameters. If it is less, use the first scaling factor; otherwise, use the second scaling factor. If the operating state is the lane centering state, use the third scaling factor; Wherein, the first scaling factor is less than the second scaling factor, and the third scaling factor is less than the second scaling factor.

2. The vehicle lateral motion control method according to claim 1, characterized in that, The step of calculating the feedback steering angle based on the LQR control method includes: determining the operating state based on the turn signal and the enable signal for the lane change or lane centering function.

3. The vehicle lateral motion control method according to claim 1, characterized in that, The step of calculating the feedforward rotation angle based on the preview tracking control includes: The target lane boundary line is obtained from the feature point coordinate sequence or the curve coefficient of the fitted curve based on the camera; wherein, the target lane boundary line includes the boundary lines of the two lanes in a lane: the current lane, the left lane and the right lane. Calculate the lane centerline based on the target lane boundary line; Select a preview point in front of the lane centerline; and, The feedforward steering angle is calculated based on the vehicle wheelbase, the relative distance between the aiming point and the vehicle, and the deflection angle between the aiming point and the vehicle.

4. The vehicle lateral motion control method according to claim 3, characterized in that, The step of selecting a preview point in front of the lane centerline includes: selecting the preview point based on the current vehicle speed and the curvature of the lane centerline; the faster the vehicle speed, the farther the preview point; the greater the curvature of the lane centerline, the closer the preview point.

5. The vehicle lateral motion control method according to claim 1, characterized in that, The steps for calculating the feedback rotation angle based on the LQR algorithm include: The lateral deviation e1 is obtained by limiting the original lateral deviation and its change slope after acquiring the original lateral deviation of the vehicle relative to the center line of the lane based on the camera. The vehicle's orientation angle deviation e2 relative to the lane centerline is obtained based on the camera. The rate of change of the vehicle's lateral deviation relative to the lane centerline is obtained by multiplying the vehicle speed and e2. 1; and, The rate of change of the vehicle's orientation angle deviation relative to the lane centerline is obtained by subtracting the yaw rate from the product of vehicle speed and lane centerline curvature.

2.

6. The vehicle lateral motion control method according to claim 1, characterized in that, The step of calculating the actual steering wheel angle based on the original value of the front wheel steering angle includes: The first steering wheel angle control value is calculated using the original value of the front wheel angle and the current steering wheel angle. If both the amplitude and rate of change of the first steering wheel angle control quantity are within a preset range, the actual steering wheel angle is calculated directly using the first steering wheel angle control quantity; and... If the amplitude or rate of change of the first steering wheel angle control quantity is not within the preset range, the first steering wheel angle control quantity is limited to obtain the second steering wheel angle control quantity, and the actual steering wheel angle is calculated based on the second steering wheel angle control quantity.

7. The vehicle lateral motion control method according to claim 1, characterized in that, The vehicle lateral motion control method also includes: The system determines whether to activate the lateral control function based on the vehicle status, external input commands, and environmental perception status. If the determination result is "activated", then the actual steering wheel angle is output to control the vehicle's movement; and, If the determination result is "not activated", then the actual steering wheel angle will not be output.

8. A vehicle lateral motion control system, characterized in that, The vehicle lateral motion control system includes: LQR feedback calculation module: used to calculate feedback steering angle based on LQR algorithm; wherein, the control deviation in the LQR algorithm is: lateral deviation of vehicle relative to lane centerline, rate of change of lateral deviation of vehicle relative to lane centerline, directional angle deviation of vehicle relative to lane centerline, and rate of change of directional angle deviation of vehicle relative to lane centerline. The pre-aiming pure tracking feedforward module calculates the feedforward rotation angle based on pre-aiming tracking control; the feedforward rotation angle is calculated based on kinematic geometry. Output module: used to obtain the original value of the front wheel steering angle by adding the feedback steering angle and the feedforward steering angle; and used to calculate the actual steering wheel angle based on the original value of the front wheel steering angle; The steps for calculating the feedback rotation angle based on the LQR algorithm include: The scaling factor is determined based on the operating state and the vehicle's kinematic parameters, wherein the operating state is one of the following states: lane centering state and instructed lane change state; and, The feedback angle is calculated by multiplying and summing the feedback gain matrix, the scaling factor, and the control deviation in the LQR algorithm. The step of determining the scaling factor based on the operating state and the vehicle's kinematic parameters includes: If the operating state is the lane change command state, determine whether the lateral position deviation of the current vehicle is less than a preset deviation threshold based on the kinematic parameters. If it is less, use the first scaling factor; otherwise, use the second scaling factor. If the operating state is the lane centering state, use the third scaling factor; Wherein, the first scaling factor is less than the second scaling factor, and the third scaling factor is less than the second scaling factor.

Images