Lane keeping angle planning method and system

Optimizing the steering wheel angle through camera perception processing and particle swarm optimization algorithm, solving the problem of large calculation and low safety of lane centering function in ADAS, achieving efficient and reliable lane-keeping angle planning, and improving vehicle driving comfort and safety .

CN119078822AActive Publication Date: 2024-12-06UNITED AUTOMOTIVE ELECTRONICS SYST
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202411241386.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-05
Publication Date
2024-12-06
Estimated Expiration
2044-09-05

AI Technical Summary

Technical Problem

The existing lane centering function in the ADAS field is due to the problem of large computing volume, low security and reliability. It fails to fully utilize the information captured by the front camera, and the complexity and computing requirements of the model prediction control algorithm are high.

Method used

The lane information is obtained through camera perception processing, a vehicle motion model is established, and the steering wheel angle is optimized using particle swarm optimization algorithm, combining polynomial fitting and pure kinematics/two-degree-of-freedom dynamics model to reduce the calculation amount and improve safety.

Benefits of technology

It significantly reduces the calculation amount of lane keeping angle planning, improves vehicle driving comfort and safety, meets the computing needs of ADAS controllers, and ensures safety in various situations through redundant strategies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119078822A_ABST
    Figure CN119078822A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of vehicle control, in particular to a lane keeping turning angle planning method and system. The method comprises the following steps: step S1, sensing and processing lane information through a camera; s2, establishing a vehicle motion model used for outputting a predicted value of vehicle position change based on the input steering wheel angle; s3, based on the vehicle motion model, performing optimization calculation on the steering wheel turning angle for lane keeping by adopting a particle swarm optimization algorithm so as to obtain an optimal steering wheel target turning angle; and S4, controlling the steering of the vehicle according to the optimal steering wheel target turning angle obtained through calculation. According to the method, the information of the center line of the road ahead is captured through the camera, the steering function of the vehicle is planned through the particle swarm optimization algorithm, the calculated amount is remarkably reduced, and therefore the driving comfort and safety of the vehicle can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of vehicle control technology, and more specifically, to a lane keeping angle planning method and system. Background Art

[0002] The Advanced Driving Assistance System (ADAS) uses various sensors installed on the vehicle to perceive and collect information on dynamic and static objects in the surrounding environment in real time, and combines it with navigation map data for comprehensive calculations and analysis, thereby alerting the driver to potential risks in advance and significantly improving driving comfort and safety.

[0003] Among the many functions of ADAS, lane centering is an important part of ADAS, ensuring that the vehicle stays centered in the autonomous driving lane. This function achieves precise control of the vehicle's driving direction by calculating and adjusting the steering wheel angle in real time.

[0004] Figure 1 The schematic diagram of the lane keeping algorithm principle of a lane centering function in the prior art is disclosed, such as Figure 1 As shown in the figure, the main method used by the lane centering function is to preview the target position at a certain distance ahead, and adjust the current steering wheel angle based on the closed-loop control concept to ensure that the vehicle is stable in the center of the lane. The operation of the lane keeping algorithm depends on a number of key input information, including real-time road condition monitoring data from the front camera, steering wheel angle information, vehicle speed, and steering wheel torque information. By comprehensively using this information and combining it with the judgment logic of the lane keeping enable state, the lane keeping algorithm can accurately calculate and output the target wheel angle value, thereby achieving precise control of the vehicle's driving trajectory and ensuring the stable implementation of the lane keeping function.

[0005] However, although this method is simple and easy to use, it fails to fully utilize all the information captured by the image sensor of the front camera. In fact, the image sensor can capture the position information of all points on the center line of the lane within a certain distance ahead, but in the actual operation of the lane centering function, only the information of the preview point is used, and the position data of other points on the center line are not used.

[0006] To overcome the above problems, some existing algorithms use model predictive control methods. This method optimizes the vehicle steering operation throughout the process by predicting the lane line equation within a certain distance ahead. However, the complexity of the model predictive control algorithm is closely related to the length of the future prediction time, and long-term prediction will result in a huge amount of calculation. In addition, model predictive control involves solving quadratic programming problems, and the existence of its solution cannot always be guaranteed. Therefore, more complex processing measures need to be taken in practical applications to ensure safety, which places high demands on the computing power of the ADAS controller.

[0007] Therefore, the ADAS field currently still lacks an efficient and reliable lane keeping angle planning method for lane centering function. Summary of the invention

[0008] The purpose of the present invention is to provide a lane keeping angle planning method and system to solve the problems of large amount of calculation and low safety and reliability in lane keeping angle planning in the prior art.

[0009] In order to achieve the above object, the present invention provides a lane keeping angle planning method, comprising the following steps:

[0010] Step S1, sensing and processing lane information through a camera;

[0011] Step S2, establishing a vehicle motion model for outputting a predicted value of a vehicle position change based on an input steering wheel angle;

[0012] Step S3, based on the vehicle motion model, using a particle swarm optimization algorithm to optimize the steering wheel angle for lane keeping, so as to obtain an optimal steering wheel target angle;

[0013] Step S4, controlling the steering of the vehicle according to the calculated optimal steering wheel target angle.

[0014] In some embodiments, the step S1 further comprises:

[0015] The lane information is sensed and processed to obtain the coordinate point data of the lane centerline.

[0016] In some embodiments, the step S1 further comprises:

[0017] According to the obtained coordinate point data of the lane center line, the curve equation of the lane center line is obtained by using the polynomial fitting method.

[0018] In some embodiments, the vehicle motion model of step S2 includes a pure kinematic model and / or a vehicle two-degree-of-freedom dynamics model:

[0019] The pure kinematic model describes the behavior of the vehicle based on the vehicle's speed, steering angle and motion path;

[0020] The two-degree-of-freedom dynamics model of the vehicle describes the vehicle behavior based on the force analysis on the lateral degree of freedom and the longitudinal degree of freedom of the vehicle.

[0021] In some embodiments, the vehicle motion model is judged and switched according to the vehicle speed:

[0022] When the vehicle speed is less than a preset threshold, a pure kinematic model is used;

[0023] When the vehicle speed is greater than or equal to a preset threshold, the vehicle two-degree-of-freedom dynamic model is used.

[0024] In some embodiments, the particle swarm optimization algorithm in step S3 includes a particle swarm optimization algorithm with adaptive weights and a particle swarm optimization algorithm with activation factors.

[0025] In some embodiments, the particle swarm optimization algorithm in step S3 further comprises the following steps:

[0026] Input the steering wheel angle as the particle coordinates;

[0027] The performance parameters of lane centering are set as the global optimization objective function and the particle swarm optimization algorithm is executed;

[0028] Obtain the optimal steering wheel angle after optimization calculation.

[0029] In some embodiments, the lane centering performance parameters include:

[0030] The heading angle deviation between the vehicle's driving direction and the lane line direction;

[0031] The lateral deviation between the vehicle center position and the lane line;

[0032] Front wheel steering angle change rate;

[0033] The magnitude of the front wheel steering angle.

[0034] In some embodiments, the particle swarm optimization algorithm in step S3 further comprises:

[0035] The loss of several lane centering performance parameters is taken as the cost, and several costs are combined according to different weights to obtain the optimization target.

[0036] In some embodiments, the particle swarm optimization algorithm in step S3 further comprises:

[0037] The estimated value is obtained as the initial value of the particle through the curve equation of the lane centerline and the current motion state of the vehicle.

[0038] In some embodiments, after step S3, the method further includes:

[0039] Conduct rationality checks on optimization calculation results;

[0040] If the check fails, the particle swarm optimization algorithm is re-performed until the optimization calculation result meets the rationality check requirements or the number of searches reaches the preset limit.

[0041] In some embodiments, if the rationality check fails and the particle swarm optimization algorithm cannot search for a reasonable solution, a backup algorithm is triggered to obtain an optimization calculation result.

[0042] In some embodiments, the backup algorithm will combine the search results of the particle swarm optimization algorithm at the previous moment and the current result calculated using the specified tracking algorithm to obtain the final optimization calculation result, and the final optimization calculation result includes the optimal steering wheel target angle at the current moment.

[0043] In some embodiments, the specified tracking algorithm is a pure tracking algorithm, comprising the following steps:

[0044] Calculate the preview point range based on the road centerline curvature and vehicle speed;

[0045] Searching for a preview distance with the minimum cost within the preview distance range, wherein the cost includes a deviation between the predicted vehicle trajectory and the lane centerline and a rate of change of the steering wheel angle;

[0046] The preview point with the lowest cost is selected as the final preview point, and the steering wheel angle is calculated based on the geometric relationship of the vehicle's motion so that the vehicle can move through the final preview point.

[0047] In order to achieve the above object, the present invention provides a lane keeping angle planning system, which at least includes an advanced driver assistance system:

[0048] The advanced driver assistance system is used to execute the lane keeping angle planning method as described above.

[0049] The present invention proposes a lane keeping angle planning method and system, which captures the information of the center line of the road ahead through a camera and plans the steering function of the vehicle using a particle swarm optimization algorithm, which significantly reduces the amount of calculation and also adopts a redundant strategy to ensure the safety of the vehicle in various calculation situations, thereby helping to improve the driving comfort and safety of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The above and other features, properties and advantages of the present invention will become more apparent through the following description in conjunction with the accompanying drawings and embodiments, in which the same reference numerals always represent the same features, wherein:

[0051] Figure 1 A schematic diagram of the lane keeping algorithm principle of a lane centering function in the prior art is disclosed;

[0052] Figure 2 A step diagram of a lane keeping corner planning method according to an embodiment of the present invention is disclosed;

[0053] Figure 3 A functional block diagram of a lane keeping angle planning method according to an embodiment of the present invention is disclosed;

[0054] Figure 4 A schematic diagram of a pure kinematic model according to an embodiment of the present invention is disclosed;

[0055] Figure 5 A method flow chart of a particle swarm algorithm according to an embodiment of the present invention is disclosed;

[0056] Figure 6 A flow chart of a method for checking algorithm rationality according to an embodiment of the present invention is disclosed;

[0057] Figure 7 A schematic diagram of a particle swarm search result of a backup algorithm according to an embodiment of the present invention is disclosed;

[0058] Figure 8 A flow chart of a pure tracking method according to an embodiment of the present invention is disclosed;

[0059] Figure 9a A schematic diagram of a vehicle trajectory of a first simulation planned by a particle swarm algorithm according to an embodiment of the present invention is disclosed;

[0060] Figure 9b A schematic diagram of steering wheel angle planning for the first simulation according to a particle swarm algorithm planning according to an embodiment of the present invention is disclosed;

[0061] Fig.10a A schematic diagram of a vehicle trajectory of a second simulation planned by a particle swarm algorithm according to an embodiment of the present invention is disclosed;

[0062] Fig.10b A schematic diagram of steering wheel angle planning for a second simulation using a particle swarm algorithm according to an embodiment of the present invention is disclosed. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the invention and are not used to limit the invention.

[0064] Figure 2 A method for lane keeping corner planning according to an embodiment of the present invention is disclosed. Figure 3 A functional block diagram of a lane keeping corner planning method according to an embodiment of the present invention is disclosed. Figure 2 and Figure 3 As shown, a lane keeping angle planning method proposed by the present invention comprises the following steps:

[0065] Step S1, sensing and processing lane information through a camera;

[0066] Step S2, establishing a vehicle motion model for calculating and outputting a predicted value of a vehicle position change based on an input steering wheel angle;

[0067] Step S3, based on the vehicle motion model, using a particle swarm optimization algorithm to optimize the steering wheel angle for lane keeping, so as to obtain an optimal steering wheel target angle;

[0068] Step S4, planning and controlling the steering of the vehicle according to the calculated optimal steering wheel target angle.

[0069] A lane keeping turning angle planning method proposed in the present invention captures the information of the center line of the road ahead through a camera and plans the steering function of the vehicle, which significantly reduces the amount of calculation and also adopts a redundant strategy to ensure the safety of the vehicle in various calculation situations.

[0070] These steps will be described in detail below. It should be understood that within the scope of the present invention, the above-mentioned technical features of the present invention and the technical features specifically described below (such as embodiments) can be combined and related to each other to form a preferred technical solution.

[0071] Step S1, sensing and processing lane information through a camera.

[0072] This step senses, extracts and processes the lane information to further obtain the coordinate point data of the lane centerline.

[0073] The curve equation of the lane centerline is calculated based on the acquired lane centerline coordinate point data.

[0074] In the process of obtaining the lane line equation, a polynomial fitting method is usually used. By adjusting the degree and coefficient of the polynomial, the fitted curve equation can be closest to the actual lane center line.

[0075] Among them, cubic polynomial fitting is one of the commonly used and effective methods, but according to the specific situation, other suitable fitting methods can also be flexibly selected for processing, for example, a higher degree polynomial or piecewise fitting method may be more applicable.

[0076] Step S2, establishing a vehicle motion model for calculating and outputting a predicted value of the vehicle position change based on the input steering wheel angle.

[0077] In this embodiment, the vehicle motion model includes a pure kinematics model and / or a vehicle two-degree-of-freedom dynamics model:

[0078] The pure kinematic model describes the behavior of the vehicle based on the vehicle's speed, steering angle and motion path;

[0079] The two-degree-of-freedom dynamics model of the vehicle describes the vehicle behavior based on the force analysis on the lateral degree of freedom and the longitudinal degree of freedom of the vehicle.

[0080] The selection of the vehicle motion model should be based on specific calculation requirements. In the initial stage, a relatively simple pure kinematic model can be used as a preliminary analysis tool, that is, the first model. As the vehicle speed increases, in order to more accurately simulate the dynamic behavior of the vehicle, you can switch to the vehicle two-degree-of-freedom dynamics model as an advanced analysis tool, that is, the second model. The switching standard between the two is based on the vehicle speed. Specifically, when the vehicle speed is low and less than the preset threshold, the first model is used; when the vehicle speed reaches the preset threshold, the second model is used.

[0081] For the above two vehicle motion models, if the vehicle speed is determined and the steering wheel angle sequence is input, the change sequence of the vehicle position can be accurately calculated.

[0082] Figure 4 The schematic diagram of the pure kinematic model according to an embodiment of the present invention is disclosed, as shown in FIG. Figure 4 The pure kinematic model shown here focuses on the position and velocity of the vehicle without involving the calculation of specific dynamics and forces. The pure kinematic model describes the behavior of the vehicle based on its geometric and kinematic properties.

[0083] In this pure kinematic model, the motion of the vehicle can be described by the following equations, which are expressed as follows:

[0084]

[0085] Among them, s represents the speed of the vehicle, φ represents the steering angle, L represents the distance between the front wheel and the rear wheel, θ represents the heading angle, and (x, y) is the coordinate of the vehicle.

[0086] In this embodiment, the two-degree-of-freedom dynamics model of the vehicle takes into account the longitudinal and lateral dynamics of the vehicle, and can provide a more accurate description of the vehicle behavior, especially considering the dynamic characteristics of the vehicle and the influence of external forces.

[0087] The vehicle two-degree-of-freedom dynamics model usually includes the following two degrees of freedom:

[0088] Lateral degrees of freedom, used to describe the vehicle's steering and lateral forces, including lateral acceleration, yaw rate, etc.;

[0089] The yaw degree of freedom is used to describe the steering motion of the vehicle, including the yaw angular velocity.

[0090] The two-degree-of-freedom dynamic model of the vehicle ignores the role of the suspension, simplifies the entire vehicle into two wheels, assumes that the tire side slip characteristics are linear, and ignores the longitudinal drive or resistance, assuming that the longitudinal vehicle speed is constant.

[0091] Step S3, based on the vehicle motion model, a particle swarm optimization algorithm is used to optimize and calculate the steering wheel angle for lane keeping, so as to obtain an optimal steering wheel target angle.

[0092] In this embodiment, the step S3 further includes the following steps:

[0093] Input the steering wheel angle as the particle coordinates;

[0094] The performance parameters of lane centering are set as the global optimization objective function and the particle swarm optimization algorithm is executed;

[0095] Obtain the optimal steering wheel angle after optimization calculation.

[0096] Figure 5 A method flow chart of a particle swarm algorithm according to an embodiment of the present invention is disclosed. Figure 5 As shown, the particle swarm algorithm further comprises the following steps:

[0097] Initialize particle swarm;

[0098] Calculate the vehicle's driving trajectory based on the vehicle motion model;

[0099] Calculate the performance of each particle;

[0100] Update the individual optimal position and global optimal position of particles;

[0101] Update the particle's velocity and position;

[0102] Determine whether the performance condition is met or the maximum number of iterations is reached;

[0103] If yes, the global optimal solution is output; if no, the vehicle's driving trajectory is recalculated.

[0104] In the lane centering application, the subsequent steering wheel angle input is used as the coordinates of the particle, and the global optimal goal is set to the performance of lane centering.

[0105] The performance parameters of lane centering include:

[0106] The heading angle deviation between the vehicle's driving direction and the lane line direction;

[0107] The lateral deviation between the vehicle center position and the lane line;

[0108] Front wheel steering angle change rate;

[0109] The magnitude of the front wheel steering angle.

[0110] Taking the loss of the lane centering performance parameter as the cost, the above costs are combined according to different weights to obtain the target to be optimized;

[0111] Among them, the weight can be adjusted according to different vehicle speeds and centerline curvatures.

[0112] According to the above particle swarm algorithm, the sequence of steering wheel angles will be calculated, and the first value will be taken as the current target steering wheel angle.

[0113] Furthermore, the particle swarm initialization step of the particle swarm algorithm includes:

[0114] The estimated value is obtained as the initial value of the particle through calculation of the curve equation of the lane centerline and the current motion state of the vehicle.

[0115] The initialization of particles has a significant impact on the particle swarm algorithm. By combining the curve equation of the lane centerline and the current motion state of the vehicle with the basic motion model of the vehicle, positions where turning is impossible can be excluded, thereby accelerating the convergence process of particles, which is helpful for the subsequent optimization search process.

[0116] It should be clear that the core of the present invention is the particle swarm algorithm, and the algorithm scope it covers should also include its various optimization deformation algorithms, such as the particle swarm algorithm with adaptive weights and the particle swarm algorithm with activation factors.

[0117] At the same time, as an optimization search algorithm, the above particle swarm algorithm may not produce acceptable search results in some specific cases. Therefore, it is necessary to verify the rationality of the results of the particle swarm algorithm.

[0118] Figure 6 A flow chart of a method for checking algorithm rationality according to an embodiment of the present invention is disclosed. Figure 6As shown in the figure, after the particle swarm algorithm searches out the optimization calculation results, the rationality of the optimization calculation results is further checked;

[0119] If the check fails, the particle swarm optimization algorithm is re-performed until the optimization calculation result meets the rationality check requirements or the number of searches reaches the preset limit.

[0120] In this embodiment, the rationality check includes:

[0121] Whether the amplitude of the steering wheel angle exceeds the limit;

[0122] Whether the steering wheel change rate exceeds the limit;

[0123] Whether the maximum value, average value, and standard deviation of the deviation between the vehicle position and the centerline position calculated by the model exceed the limit.

[0124] If all the above rationality checks fail, the particle swarm algorithm will be re-run until a reasonable result is found or the search number reaches the limit.

[0125] If the rationality check fails and the particle swarm optimization algorithm cannot search for a reasonable solution, the backup algorithm is triggered to obtain the optimization calculation result.

[0126] Furthermore, the backup algorithm will combine the search results of the particle swarm optimization algorithm at the previous moment and the current result calculated by the specified tracking algorithm to obtain the final optimization calculation result, and the final optimization calculation result includes the optimal steering wheel target angle at the current moment.

[0127] Specifically, the backup algorithm combines the results of the second step of the search results at the previous moment with the current results calculated using the pure tracking algorithm to obtain the solution of the backup algorithm. This solution only contains the target corner at the current moment.

[0128] Figure 7 A schematic diagram of a particle swarm search result of a backup algorithm according to an embodiment of the present invention is disclosed, such as Figure 7 As shown in the figure, in the particle swarm optimization algorithm operation, the search result obtained at the last moment is regarded as the optimal solution, which is presented in the form of an array, specifically including 3 elements, which represent the corner required to be executed at the current time node at the last moment, the corner required to be executed at the next time node at the last moment, and the corner required to be executed at the next time node at the last moment. Therefore, the second element in the optimal solution, the corner required to be executed at the next time node at the last moment, corresponds to the corner required to be executed at the current moment, which is the result of the second step of the search result of the particle swarm optimization algorithm at the last moment.

[0129] In this embodiment, the designated tracking algorithm is a pure tracking algorithm, which is a pure kinematic control method with relatively simple functions and can be used as a redundant safety backup algorithm.

[0130] The pure tracking algorithm selects a point on the reference trajectory that is a certain preview distance away from the vehicle's current position as the preview point, and controls the vehicle to drive along the trajectory toward the preview point by adjusting the steering wheel angle. In actual applications, the preview distance will be optimized to achieve better control effects.

[0131] Figure 8 A flow chart of a pure tracking method according to an embodiment of the present invention is disclosed, wherein the specified tracking algorithm is a pure tracking algorithm, comprising the following steps:

[0132] Calculate the preview point range based on the road centerline curvature and vehicle speed;

[0133] Searching for a preview distance with the minimum cost within the preview distance range, wherein the cost includes a deviation between the predicted vehicle trajectory and the lane centerline and a rate of change of the steering wheel angle;

[0134] The preview point with the lowest cost is selected as the final preview point, and the steering wheel angle is calculated based on the geometric relationship of the vehicle's motion so that the vehicle can move through the final preview point.

[0135] Step S4, planning and controlling the steering of the vehicle according to the calculated optimal steering wheel target angle.

[0136] Figure 9a A schematic diagram of a vehicle trajectory of the first simulation according to a particle swarm algorithm planning according to an embodiment of the present invention is disclosed, such as Figure 9a As shown in the figure, the dots simulate the location of the lane centerline obtained by the camera. The vehicle speed is a fixed value, and the solid line is the planned vehicle trajectory.

[0137] Figure 9b Revealed and Figure 9a Correspondingly, the schematic diagram of the steering wheel angle planning for the first simulation using the particle swarm algorithm is shown in the figure below. Figure 9b The corner planning shown in the figure is calculated based on the vehicle model, and the corresponding planning path is Figure 9a It can be seen that the planned path can closely fit the center line of the lane, and the change of the steering wheel angle is also relatively smooth, which helps to improve the driving comfort of the vehicle.

[0138] Fig.10a A schematic diagram of a vehicle trajectory of a second simulation planned by a particle swarm algorithm according to an embodiment of the present invention is disclosed. Fig.10bThe schematic diagram of the second simulation of the steering wheel angle planning of the particle swarm algorithm planning according to an embodiment of the present invention is disclosed. The dots in the figure represent the position points of the lane center line captured by the simulated camera. The vehicle speed is set to a constant value, and the solid line represents the planned vehicle trajectory. There is a certain deviation between the simulated vehicle and the lane line (the simulation sets a 1m deviation). It can be seen that the planned trajectory can smoothly (within 3s) gradually approach the lane center line, thereby achieving tracking of the lane center line.

[0139] The computational requirements of the algorithm can be met on existing ADAS controllers, which is significantly reduced compared to model predictive control (MPC).

[0140] Furthermore, the structure of the PSO algorithm allows the number of particles to be grouped, thus enabling parallel computation, which helps to reduce the overall computational time of the algorithm.

[0141] Furthermore, the present invention also proposes a lane keeping angle planning system, which at least includes an advanced driver assistance system:

[0142] The advanced driver assistance system is used to execute the lane keeping angle planning method mentioned above.

[0143] The specific details of the lane keeping angle planning system have been explained in detail in the previous lane keeping angle planning method and will not be repeated here.

[0144] The present invention proposes a lane keeping angle planning method and system, which captures the information of the center line of the road ahead through a camera and plans the steering function of the vehicle using a particle swarm optimization algorithm, which significantly reduces the amount of calculation and also adopts a redundant strategy to ensure the safety of the vehicle in various calculation situations, thereby helping to improve the driving comfort and safety of the vehicle.

[0145] Although the above methods are illustrated and described as a series of actions for simplicity of explanation, it should be understood and appreciated that these methods are not limited by the order of the actions, because according to one or more embodiments, some actions may occur in a different order and / or concurrently with other actions from those illustrated and described herein or not illustrated and described herein but understandable to those skilled in the art.

[0146] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0147] Those skilled in the art will appreciate that information, signals, and data may be represented using any of a variety of different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips cited throughout the above description may be represented by voltage, current, electromagnetic waves, magnetic fields or magnetic particles, optical fields or optical particles, or any combination thereof.

[0148] Those skilled in the art will further appreciate that the various illustrative logic blocks, modules, circuits, and algorithm steps described in conjunction with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or a combination of the two. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps are generally described above in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system. The technician may implement the described functionality in different ways for each specific application, but such implementation decisions should not be interpreted as resulting in a departure from the scope of the present invention.

[0149] The above embodiments are provided for persons familiar with the art to implement or use the present invention. Personnel familiar with the art can make various modifications or changes to the above embodiments without departing from the inventive concept of the present invention. Therefore, the protection scope of the present invention is not limited to the above embodiments, but should be the maximum scope of the innovative features mentioned in the claims.

Claims

1. A lane keeping angle planning method, characterized in that: The following steps are involved: Step S1, sensing and processing lane information through a camera; Step S2, establishing a vehicle motion model for outputting a predicted value of a vehicle position change based on an input steering wheel angle; Step S3, based on the vehicle motion model, using a particle swarm optimization algorithm to optimize the steering wheel angle for lane keeping, so as to obtain an optimal steering wheel target angle; Step S4, controlling the steering of the vehicle according to the calculated optimal steering wheel target angle.

2. The lane keeping angle planning method according to claim 1, characterized in that: The step S1 further comprises: The lane information is sensed and processed to obtain the coordinate point data of the lane centerline.

3. The lane keeping angle planning method according to claim 2, characterized in that: The step S1 further comprises: According to the obtained coordinate point data of the lane center line, the curve equation of the lane center line is obtained by using the polynomial fitting method.

4. The lane keeping angle planning method according to claim 1, characterized in that: The vehicle motion model in step S2 includes a pure kinematics model and / or a vehicle two-degree-of-freedom dynamics model: The pure kinematic model describes the behavior of the vehicle based on the vehicle's speed, steering angle and motion path; The two-degree-of-freedom dynamics model of the vehicle describes the vehicle behavior based on the force analysis on the lateral degree of freedom and the longitudinal degree of freedom of the vehicle.

5. The lane keeping angle planning method according to claim 4, characterized in that: The vehicle motion model is judged and switched according to the vehicle speed: When the vehicle speed is less than a preset threshold, a pure kinematic model is used; When the vehicle speed is greater than or equal to a preset threshold, the vehicle two-degree-of-freedom dynamic model is used.

6. The lane keeping angle planning method according to claim 1, characterized in that: The particle swarm optimization algorithm in step S3 includes a particle swarm optimization algorithm with adaptive weights and a particle swarm optimization algorithm with activation factors.

7. The lane keeping angle planning method according to claim 1, characterized in that: The step S3 further comprises the following steps: Input the steering wheel angle as the particle coordinates; The performance parameters of lane centering are set as the global optimization objective function and the particle swarm optimization algorithm is executed; Obtain the optimal steering wheel angle after optimization calculation.

8. The lane keeping angle planning method according to claim 7, characterized in that: The performance parameters of lane centering include: The heading angle deviation between the vehicle's driving direction and the lane line direction; The lateral deviation between the vehicle center position and the lane line; Front wheel steering angle change rate; The magnitude of the front wheel steering angle.

9. The lane keeping angle planning method according to claim 7 or 8, characterized in that: The particle swarm optimization algorithm in step S3 further comprises: The losses of several lane centering performance parameters are taken as costs, and the several costs are combined according to different weights to obtain an optimization target.

10. The lane keeping angle planning method according to claim 7, characterized in that: The particle swarm optimization algorithm in step S3 further comprises: The estimated value is obtained as the initial value of the particle through the curve equation of the lane centerline and the current motion state of the vehicle.

11. The lane keeping angle planning method according to claim 1, characterized in that: After step S3, the method further comprises: Conduct rationality checks on optimization calculation results; If the check fails, the particle swarm optimization algorithm is re-performed until the optimization calculation result meets the rationality check requirements or the number of searches reaches the preset limit.

12. The lane keeping angle planning method according to claim 11, characterized in that: If the rationality check fails and the particle swarm optimization algorithm cannot search for a reasonable solution, the backup algorithm is triggered to obtain the optimization calculation result.

13. The lane keeping angle planning method according to claim 12, characterized in that: The backup algorithm combines the search results of the particle swarm optimization algorithm at the previous moment and the current result calculated by the specified tracking algorithm to obtain the final optimization calculation result, which includes the optimal steering wheel target angle at the current moment.

14. The lane keeping angle planning method according to claim 13, characterized in that: The specified tracking algorithm is a pure tracking algorithm, comprising the following steps: Calculate the preview point range based on the road centerline curvature and vehicle speed; Searching for a preview distance with the minimum cost within the preview distance range, wherein the cost includes a deviation between the predicted vehicle trajectory and the lane centerline and a rate of change of the steering wheel angle; The preview point with the lowest cost is selected as the final preview point, and the steering wheel angle is calculated based on the geometric relationship of the vehicle's motion so that the vehicle can move through the final preview point.

15. A lane keeping angle planning system, characterized in that: Includes at least advanced driver assistance systems: The advanced driver assistance system is used to execute the lane keeping angle planning method as described in any one of claims 1-14.

Citation Information

Patent Citations

  • Four-wheel steering model prediction control method based on data driving

    CN111055921A

  • Intelligent vehicle lane centering keeping method

    CN112537303A

  • Transverse stability control method of unmanned vehicle

    CN114162110A

  • Lane keeping control method based on multi-point preview

    CN116118725A

  • Self-adaptive path tracking method and system based on PSO-LQR algorithm

    CN117970796A