Lane centering driving trajectory decision method, device and storage medium

The curvature equation is obtained by the lane line recognition and tracking module. The tracking trajectory for the next frame is calculated by combining the discreteness and the curve radius. The steering angle control is calculated by the LQR controller, which solves the stability problem of lane centering control on the L2 platform and improves the trajectory stability and adaptability.

CN118651220BActive Publication Date: 2024-11-22SHENZHEN ANNGIC TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411140713.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-20
Publication Date
2024-11-22
Estimated Expiration
2044-08-20

AI Technical Summary

Technical Problem

Existing technologies suffer from reduced centering performance in lane keeping control, especially in situations involving merging, splitting, or unclear lane markings, making stable deployment on L2 platforms difficult.

Method used

The lane line curvature equation is obtained by the lane line recognition and tracking module. The tracking trajectory for the next frame is calculated by combining the discreteness, curve radius and length. The LQR controller is used to calculate the steering angle control quantity to improve the trajectory stability.

Benefits of technology

It improves the stability of lane centering control, adapts to scenarios of traffic divergence and merging, and vehicles cutting in front and crossing the line, reduces the risk of unexpected steering, and enhances performance in situations where the single-sided line is unclear.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118651220B_ABST
    Figure CN118651220B_ABST
Patent Text Reader

Abstract

The present invention discloses a lane centering driving trajectory decision method, comprising the following steps: A, a lane line recognition and tracking module recognizes and tracks the lane line in each frame image captured by a vehicle-mounted camera, and obtains the curvature equation and lane line length of each lane line in each frame image; B, a centering driving trajectory decision module calculates the discreteness of the lane line and the previous frame tracking trajectory line, as well as the curve radius and lane line length of the lane line according to the curvature equation, and obtains the next frame tracking trajectory; C, a turning angle instruction calculation module sends the next frame tracking trajectory obtained in step B to the LQR controller module to calculate the turning angle control amount relative to the previous frame tracking trajectory. The present invention can improve the deficiencies of the prior art and make the lane centering control more stable.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of vehicle assisted driving technology, and in particular to a lane centering driving trajectory decision method, device and storage medium. Background Art

[0002] In the field of intelligent driving, lane keeping is responsible for centering control based on the lane lines detected by the lane line detection module. It is highly dependent on the results detected by the lane line detection module. When there is a merge or a vehicle on one side crosses the line or the lane line is unclear, the centering performance will be significantly reduced or even unexpected steering will occur. Later, there were methods based on model predictive control or planning to achieve lane keeping assistance. This type of method can take into account both comfort and safety, but it has a large amount of calculation and is mostly used for high-level assisted driving. L2-level assisted driving mostly uses low-cost computing platforms, which are difficult to deploy on L2 platforms. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a lane centering driving trajectory decision method, device and storage medium, which can solve the shortcomings of the existing technology and make the lane centering control more stable.

[0004] In order to solve the above technical problems, the technical solutions adopted by the present invention are as follows.

[0005] A lane centering driving trajectory decision method includes the following steps:

[0006] A. The lane line recognition and tracking module recognizes and tracks the lane lines in each frame image taken by the vehicle camera, and obtains the curvature equation and lane line length of each lane line in each frame image;

[0007] B. The centering driving trajectory decision module calculates the discreteness of the lane line and the previous frame tracking trajectory line, as well as the curve radius and lane line length of the lane line according to the curvature equation, and obtains the next frame tracking trajectory;

[0008] C. The turning angle command calculation module sends the next frame tracking trajectory obtained in step B to the LQR controller module to calculate the turning angle control amount relative to the previous frame tracking trajectory.

[0009] Preferably, in step A, the lane curvature equation is expressed as:

[0010] ,

[0011] Among them, c3 is the first-order differential of the lane line curvature, c2 is the lane line curvature, c1 is the lane slope, and c0 is the lateral distance of the lane line from the origin of the coordinate system.

[0012] Preferably, step B specifically includes the following steps:

[0013] B1, calculate the discreteness of the current left and right lane lines and the tracking trajectory of the previous frame. If the tracking trajectory of the next frame can be determined by the discreteness, then end step B, otherwise go to step B2;

[0014] B2, calculate the lane curve radius and determine the merging and diverging. If the curve radius can determine the tracking trajectory of the next frame, then end step B, otherwise go to step B3;

[0015] B3. Calculate the difference in length between the left and right lane lines. If the difference in length between the left and right lane lines is greater than the set threshold, take the parameters c3, c2, c1 of the longer lane line as the parameters c3, c2, c1 of the next frame tracking trajectory line, and take the average value of the left and right lane line parameters c0 as the next frame tracking trajectory line parameter c0. Otherwise, take the average value of the left and right lane line parameters as the next frame tracking trajectory line parameter.

[0016] Preferably, in step B1,

[0017] Discretize the lane line curvature direction and the tracking trajectory of the previous frame, take a point every 1m in the longitudinal distance, and take a maximum of n=50 points, and substitute it into the lane line curvature equation to obtain the lane line discrete point set.

[0018] The set of discrete points of the left lane line: L = [(XL1, YL1), (XL2, YL2)… (XL n , YL n )],

[0019] The discrete point set of the right lane line: R = [(XR1, YR1), (XR2, YR2)… (XR n , YR n )],

[0020] The set of discrete points of the previous frame tracking trajectory: C = [(XC1, YC1), (XC2, YC2)… (XC n , Y.C. n )];

[0021] Calculate the sum of the Euclidean distances of the left lane line discrete point set and the right lane line discrete point set relative to the corresponding points of the previous frame tracking trajectory discrete point set, which represents the trajectory discreteness.

[0022] Wherein, SimilVal1 is the discreteness between the left lane line and the tracking trajectory of the previous frame, SimilVal2 is the discreteness between the right lane line and the tracking trajectory of the previous frame, and step B is ended; otherwise, go to step B2.

[0023] Preferably, in step B2,

[0024] The average curve radius R is calculated based on the curvature equations of the left and right lane lines. A point is taken every 5m in the longitudinal distance for calculation, with a maximum of n=8 points. The curvature radius of the corresponding point is calculated based on the following formula. The curve radii of the 8 points are averaged to obtain the average curve radius;

[0025] ,

[0026] Among them, y' is the first-order derivative of function y with respect to x, y'' is the second-order derivative of function y with respect to x, y=f des (x);

[0027] Calculate the difference in curve radius between the left and right lane lines. If the difference in curve radius is greater than the set threshold, select the parameters c3, c2, and c1 of the lane line on the side with the larger curve radius. Record the lane width of the previous frame and the orientation of the lane line selected this time. If the selected side orientation is the left side, c0 is the difference between the lane line c0 item and half the lane width. If the selected side orientation is the right side, c0 is the sum of the lane line c0 item and half the lane width. End step B; otherwise, go to step B3.

[0028] An electronic device includes a memory and a processor, wherein the memory stores computer-executable instructions, and the processor executes the computer-executable instructions stored in the memory to implement the above-mentioned lane centering driving trajectory decision method.

[0029] A computer-readable storage medium stores computer-executable instructions, which are used to implement the lane centering driving trajectory decision method when executed by a processor.

[0030] The beneficial effects brought about by adopting the above technical solution are:

[0031] The present invention judges the trajectory discreteness, curve radius and lane line length respectively according to the left and right lane line information detected in this frame and the tracking trajectory line information determined in the previous frame; it considers the stability of the tracking trajectory during centering control from multiple dimensions, greatly improves the stability of the tracking trajectory, and makes the lane centering control more stable.

[0032] The present invention can adapt to lane line scenarios of diverging and merging, and scenarios where the vehicle in front cuts in and presses the line when controlling lane centering, greatly improving the scenario adaptability of lane centering assistance, and eliminating the risk of collision caused by unexpected or large turns.

[0033] When the perception result of one-side line is unclear and abnormal, the present invention selects a more trustworthy lane line for line filling through historical verification, and the performance will not be significantly reduced in the scenario of unclear lane line. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1It is a system principle diagram of the present invention.

[0035] Figure 2 It is the algorithm flow chart of the present invention.

[0036] Figure 3 It is a schematic diagram of the vehicle coordinate system of the present invention. DETAILED DESCRIPTION

[0037] Reference Figure 1-3 The lane centering driving trajectory decision method disclosed in the present invention comprises the following steps:

[0038] A. The lane line recognition and tracking module recognizes and tracks the lane lines in each frame image taken by the vehicle camera, and obtains the curvature equation and lane line length of each lane line in each frame image; the expression of the lane line curvature equation is:

[0039] ,

[0040] Among them, c3 is the first-order differential of the lane line curvature, c2 is the lane line curvature, c1 is the lane slope, and c0 is the lateral distance of the lane line from the origin of the coordinate system.

[0041] B. The centering driving trajectory decision module calculates the discreteness of the lane line and the previous frame tracking trajectory line, as well as the curve radius and lane line length of the lane line according to the curvature equation, and obtains the next frame tracking trajectory.

[0042] The specific calculation process includes three serial steps:

[0043] B1. Discretize the lane line curvature direction and the tracking trajectory of the previous frame, take a point every 1m in the longitudinal distance, and take a maximum of n=50 points. Substitute it into the lane line curvature equation to obtain the lane line discrete point set.

[0044] The set of discrete points of the left lane line: L = [(XL1, YL1), (XL2, YL2)… (XL n , YL n )],

[0045] The discrete point set of the right lane line: R = [(XR1, YR1), (XR2, YR2)… (XR n , YR n )],

[0046] The set of discrete points of the previous frame tracking trajectory: C = [(XC1, YC1), (XC2, YC2)… (XC n , Y.C. n )];

[0047] Calculate the sum of the Euclidean distances of the left lane line discrete point set and the right lane line discrete point set relative to the corresponding points of the previous frame tracking trajectory discrete point set, which represents the trajectory discreteness.

[0048] Among them, SimilVal1 is the discreteness of the left lane line and the tracking trajectory of the previous frame, and SimilVal2 is the discreteness of the right lane line and the tracking trajectory of the previous frame;

[0049] Calculate the discreteness difference between SimilVal1 and SimilVal2. If the discreteness difference is greater than the set threshold, select the parameters C3, C2, and C1 of the lane line with smaller discreteness and assign them to the tracking trajectory line as the parameters C3, C2, and C1 of the tracking trajectory line. The tracking trajectory line C0 takes the average value of the C0 items of the left and right lane lines, and ends step B; otherwise, go to step B2.

[0050] B2. Calculate the average curve radius R based on the curvature equations of the left and right lane lines. Take a point every 5 meters in the longitudinal distance for calculation. The maximum number of points is n=8. Calculate the curvature radius of the corresponding point based on the following formula. Average the curve radii of the 8 points to get the average curve radius.

[0051] ,

[0052] Among them, y' is the first-order derivative of function y with respect to x, y'' is the second-order derivative of function y with respect to x, y=f des (x);

[0053] Calculate the difference in curve radius between the left and right lane lines. If the difference in curve radius is greater than the set threshold, select the parameters c3, c2, and c1 of the lane line on the side with the larger curve radius. Record the lane width of the previous frame and the orientation of the lane line selected this time. If the selected side orientation is the left side, c0 is the difference between the lane line c0 item and half the lane width. If the selected side orientation is the right side, c0 is the sum of the lane line c0 item and half the lane width. End step B; otherwise, go to step B3.

[0054] B3. Calculate the difference in length between the left and right lane lines. If the difference in length between the left and right lane lines is greater than the set threshold, take the parameters c3, c2, c1 of the longer lane line as the parameters c3, c2, c1 of the next frame tracking trajectory line, and take the average value of the left and right lane line parameters c0 as the next frame tracking trajectory line parameter c0. Otherwise, take the average value of the left and right lane line parameters as the next frame tracking trajectory line parameter.

[0055] C. The turning angle command calculation module sends the next frame tracking trajectory obtained in step B to the LQR controller module to calculate the turning angle control amount relative to the previous frame tracking trajectory.

[0056] The functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0057] An electronic device is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASICs), programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0058] It should also be understood that if the above implementation is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application.

[0059] Therefore, this embodiment further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the data storage method provided in this embodiment is implemented. The computer-readable storage medium can be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc., which can store program codes.

[0060] The present invention determines the track discreteness, curve radius, and lane line length respectively based on the left and right lane line information detected in this frame and the tracking track line information determined in the previous frame; it considers the stability of the tracking track during centering control from multiple dimensions, greatly improving the stability of the tracking track and making lane centering control more stable. During lane centering control, it can adapt to the lane line scenarios of divergence and merging, and the front vehicle cutting into the line scenario, greatly improving the scene adaptability of lane centering assistance, and there will be no risk of collision caused by unexpected or large turns. In the case of an abnormal perception result of unclear unilateral lines, a more trustworthy lane line is selected for line filling through historical verification, and the performance will not be significantly reduced in scenarios where the lane lines are unclear.

[0061] In the description of the present invention, it should be understood that the terms "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0062] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A lane centering driving trajectory decision method, characterized in that The following steps are involved: A. The lane line recognition and tracking module recognizes and tracks the lane lines in each frame image taken by the vehicle camera, and obtains the curvature equation and lane line length of each lane line in each frame image; the expression of the lane line curvature equation is: , Where c3 is the first-order differential of the lane line curvature, c2 is the lane line curvature, c1 is the lane slope, and c0 is the lateral distance of the lane line from the origin of the coordinate system; B. The centering driving trajectory decision module calculates the discreteness of the lane line and the previous frame tracking trajectory line, as well as the curve radius and lane line length of the lane line according to the curvature equation, and obtains the next frame tracking trajectory; specifically, it includes the following steps: B1, calculate the discreteness of the current left and right lane lines and the tracking trajectory of the previous frame. If the tracking trajectory of the next frame can be determined by the discreteness, then end step B, otherwise go to step B2; B2, calculate the lane curve radius and determine the merging and diverging. If the curve radius can determine the tracking trajectory of the next frame, then end step B, otherwise go to step B3; B3. Calculate the difference in length between the left and right lane lines. If the difference is greater than the set threshold, take the parameters c3, c2, c1 of the longer lane line as the parameters c3, c2, c1 of the next frame tracking trajectory line. The parameter c0 of the next frame tracking trajectory line takes the average value of the parameters c0 of the left and right lane lines. Otherwise, take the average value of the parameters of the left and right lane lines as the parameter of the next frame tracking trajectory line. C. The turning angle command calculation module sends the next frame tracking trajectory obtained in step B to the LQR controller module to calculate the turning angle control amount relative to the previous frame tracking trajectory.

2. The lane centering driving trajectory decision method according to claim 1, characterized in that: In step B1, Discretize the lane line curvature direction and the tracking trajectory of the previous frame, take a point every 1m in the longitudinal distance, and take a maximum of n=50 points, and substitute it into the lane line curvature equation to obtain the lane line discrete point set. The set of discrete points of the left lane line: L = [(XL1, YL1), (XL2, YL2)… (XL n , YL n )], The discrete point set of the right lane line: R = [(XR1, YR1), (XR2, YR2)… (XR n , YR n )], The set of discrete points of the previous frame tracking trajectory: C = [(XC1, YC1), (XC2, YC2)… (XC n , Y.C. n )]; Calculate the sum of the Euclidean distances of the left lane line discrete point set and the right lane line discrete point set relative to the corresponding points of the previous frame tracking trajectory discrete point set, which represents the trajectory discreteness. Among them, SimilVal1 is the discreteness of the left lane line and the tracking trajectory of the previous frame, and SimilVal2 is the discreteness of the right lane line and the tracking trajectory of the previous frame; Calculate the discreteness difference between SimilVal1 and SimilVal2. If the discreteness difference is greater than the set threshold, select the parameters C3, C2, and C1 of the lane line with smaller discreteness and assign them to the tracking trajectory line as the parameters C3, C2, and C1 of the tracking trajectory line. The tracking trajectory line C0 takes the average value of the C0 items of the left and right lane lines, and ends step B; otherwise, go to step B2.

3. The lane centering driving trajectory decision method according to claim 2 is characterized in that: In step B2, The average curve radius R is calculated based on the curvature equations of the left and right lane lines. A point is taken every 5m in the longitudinal distance for calculation, with a maximum of n=8 points. The curvature radius of the corresponding point is calculated based on the following formula. The curve radii of the 8 points are averaged to obtain the average curve radius; , Among them, y' is the first-order derivative of function y with respect to x, y'' is the second-order derivative of function y with respect to x, y=f des (x); Calculate the difference in curve radius between the left and right lane lines. If the difference in curve radius is greater than the set threshold, select the parameters c3, c2, and c1 of the lane line on the side with the larger curve radius. Record the lane width of the previous frame and the orientation of the lane line selected this time. If the selected side orientation is the left side, c0 is the difference between the lane line c0 item and half the lane width. If the selected side orientation is the right side, c0 is the sum of the lane line c0 item and half the lane width. End step B; otherwise, go to step B3.

4. An electronic device, characterized in that: It includes a memory and a processor, the memory stores computer-executable instructions, and the processor executes the computer-executable instructions stored in the memory to implement the lane centering driving trajectory decision method as described in any one of claims 1 to 3.

5. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the lane centering driving trajectory decision method as described in any one of claims 1 to 3.