Lane line fitting method and device, equipment and storage medium

Abstract

The invention discloses a lane line fitting method and device, equipment and a storage medium, and relates to the technical field of automatic driving, and the lane line fitting method comprises the steps: obtaining a driving image, and extracting an initial lane line point set from the driving image; according to a preset lane line point set interval, a polynomial model and a lane line fitting adaptive model, an expected lane line point set is intercepted from the initial lane line point set; and fitting the expected lane line point set according to the polynomial model to obtain a target lane line equation, and completing lane line fitting. According to the method, the lane line can be accurately fitted, accurate lane information is provided for the vehicle, the transverse control effect of the automatic driving system is guaranteed, and the driving safety and stability are improved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a lane line fitting method, device, equipment and storage medium. Background Art

[0002] In autonomous driving technology, accurate identification and fitting of lane line information is crucial to the lateral control of the vehicle. Existing autonomous driving systems rely on forward-looking cameras and other sensors to extract point set information of lane lines, and fit them through methods such as polynomial arc models. However, as autonomous driving scenarios expand from highways to complex sections with small curve radii such as urban areas and rural areas, the accuracy and real-time performance of lane line fitting become more important. In order to ensure stable driving of the vehicle under complex road conditions, the lane line fitting interval needs to be adjusted according to the real-time vehicle speed and curve curvature, so as to ensure that the autonomous driving system can accurately predict and fit the lane lines.

[0003] Current autonomous driving systems usually fit lane lines through Kalman filtering, polynomial arc models, etc., and gradually fit lane lines of these small segments by dividing the lane lines into multiple small segments. However, this method performs poorly when dealing with complex scenes with large curvature of curves, and it is difficult to ensure the accuracy of lane line fitting. In addition, some solutions fail to combine dynamic information of the vehicle such as speed and curve curvature for adaptive adjustment, resulting in low fitting accuracy of the system under different road conditions.

[0004] Existing lane line fitting methods usually rely on a fixed interval point set for fitting, and do not fully consider the impact of curve curvature and vehicle speed on the fitting interval. Especially when the curve curvature changes greatly or the vehicle speed is fast, the fixed interval fitting method is likely to lead to an increase in fitting error, affecting the lateral control effect of the vehicle. In addition, traditional methods are mostly based on a single polynomial model and cannot effectively handle complex lane line shapes, such as U-shaped or arc-shaped lanes. Therefore, the existing methods are insufficient in scenarios with complex road conditions and high precision requirements, and it is urgent to improve the accuracy and stability of fitting by introducing a dynamically adjusted adaptive fitting solution. Therefore, how to accurately fit the lane line has become an urgent problem to be solved.

[0005] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0006] The purpose of this application is to provide a lane line fitting method, device, equipment and storage medium, aiming to solve the technical problem of how to accurately fit the lane line.

[0007] To achieve the above purpose, the present application proposes a lane line fitting method, which includes:

[0008] Acquire a driving image, and extract an initial lane line point set from the driving image;

[0009] Obtaining a desired lane line point set from the initial lane line point set according to a preset lane line point set interval, a polynomial model, and a lane line fitting adaptive model;

[0010] The desired lane line point set is fitted according to the polynomial model to obtain a target lane line equation, thereby completing lane line fitting.

[0011] In one embodiment, the step of obtaining a desired lane line point set from the initial lane line point set according to a preset lane line point set interval, a polynomial model, and a lane line fitting adaptive model comprises:

[0012] Obtaining a target lane line point set from the initial lane line point set according to a preset lane line point set interval;

[0013] Fitting the target lane line point set according to a polynomial model to obtain a preliminary lane line equation;

[0014] Obtaining an expected fitting point set interval according to the preliminary lane line equation, the polynomial model, the target lane line point set, and the lane line fitting adaptive model;

[0015] The expected lane line point set is obtained by intercepting the initial lane line point set according to the expected fitting point set interval.

[0016] In one embodiment, the step of obtaining the expected fitting point set interval according to the preliminary lane line equation, the polynomial model, the target lane line point set and the lane line fitting adaptive model includes:

[0017] Calculate the curvature of the curve of the fitted lane line according to the preliminary lane line equation;

[0018] When the curve curvature is greater than a preset curvature threshold, obtaining an expected fitting point set interval according to the polynomial model, the curve curvature and the lane line fitting adaptive model;

[0019] When the curve curvature is less than or equal to the preset curvature threshold, the preset lane line point set interval is used as the expected fitting point set interval.

[0020] In one embodiment, the step of obtaining the expected fitting point set interval according to the polynomial model, the curve curvature and the lane line fitting adaptive model comprises:

[0021] Obtaining the vehicle speed and the camera origin position, and inputting the curve curvature and the vehicle speed into the lane line fitting adaptive model to obtain an initial expected interval;

[0022] Calculate the fitting root mean square error according to the initial expected interval, the polynomial model and the camera origin position;

[0023] When the fitting root mean square error is greater than a preset error, adjusting the vehicle speed weight coefficient of the curve curvature and the lane line fitting adaptive model, and returning to the step of inputting the curve curvature and the vehicle speed into the lane line fitting adaptive model;

[0024] When the fitting root mean square error is less than or equal to the preset error, the initial expected interval is used as the expected fitting point set interval.

[0025] In one embodiment, the step of calculating the fitting root mean square error according to the initial expected interval, the polynomial model and the camera origin position includes:

[0026] Obtaining an original lane line point set from the initial lane line point set according to the initial expected interval;

[0027] Fitting the original lane line point set according to the polynomial model to obtain a fitted lane line equation;

[0028] Calculate a fitted lane line point set according to the fitted lane line equation;

[0029] A fitting root mean square error is calculated based on the fitting lane line point set, the original lane line point set and the camera origin position.

[0030] In one embodiment, before the step of obtaining the desired lane line point set from the initial lane line point set according to the preset lane line point set interval, the polynomial model and the lane line fitting adaptive model, the step further includes:

[0031] Get the historical lane line point set;

[0032] Calculating the historical curve curvature according to the historical lane line point set;

[0033] A lane line fitting adaptive model is constructed according to a preset critical threshold curve curvature, the historical curve curvature, a preset critical threshold vehicle speed, a preset minimum fitting interval, a preset maximum fitting interval and a vehicle speed weight coefficient.

[0034] In one embodiment, the step of fitting the desired lane line point set according to the polynomial model to obtain the target lane line equation includes:

[0035] Constructing a design matrix and a longitudinal coordinate vector according to the desired lane line point set;

[0036] Solving normal equations according to the design matrix and the longitudinal coordinate vector to obtain polynomial coefficients;

[0037] Substitute the polynomial coefficients into the polynomial model to obtain the target lane line equation.

[0038] In addition, to achieve the above purpose, the present application also proposes a lane line fitting device, the device comprising:

[0039] A data acquisition module, used to acquire a driving image and extract an initial lane line point set from the driving image;

[0040] A data processing module, used for intercepting a desired lane line point set from the initial lane line point set according to a preset lane line point set interval, a polynomial model and a lane line fitting adaptive model;

[0041] The fitting module is used to fit the desired lane line point set according to the polynomial model to obtain the target lane line equation and complete the lane line fitting.

[0042] In addition, to achieve the above-mentioned purpose, the present application also proposes a lane line fitting device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the lane line fitting method described above.

[0043] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the lane line fitting method described above are implemented.

[0044] One or more technical solutions proposed in this application have at least the following technical effects:

[0045] Acquire a driving image, and extract an initial lane line point set from the driving image; intercept the desired lane line point set from the initial lane line point set according to the preset lane line point set interval, polynomial model and lane line fitting adaptive model; fit the desired lane line point set according to the polynomial model to obtain the target lane line equation and complete lane line fitting. First, obtain real-time driving images through the vehicle's front-view camera, and extract the initial lane line point set from it using image processing technology. These point sets reflect the actual position of the lane line on the road, provide basic data for subsequent fitting, and ensure that the vehicle can perceive the road conditions ahead in real time. Then, according to the preset lane line point set interval, the polynomial model and lane line fitting adaptive model are used to filter out the desired lane line point set from the initial point set. The adaptive model dynamically adjusts the fitting interval, taking into account the real-time vehicle speed and curve curvature, so that the filtered point set is more in line with the actual road environment, improving the accuracy of fitting, and is particularly suitable for complex sections such as curves. Finally, the polynomial model is used to fit the expected lane line point set, and the polynomial coefficients are obtained by solving the normal equation to generate the target lane line equation. This equation accurately describes the shape of the lane line and can accurately fit the lane line, providing the vehicle with accurate lane information, ensuring the lateral control effect of the autonomous driving system, and improving driving safety and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0047] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0048] Figure 1 A schematic diagram of a process flow provided for the first embodiment of the lane line fitting method of the present application;

[0049] Figure 2 A schematic diagram of the target lane line equation fitting effect provided in the first embodiment of the lane line fitting method of the present application;

[0050] Figure 3 A schematic diagram of a process flow provided for the second embodiment of the lane line fitting method of the present application;

[0051] Figure 4 A schematic diagram of the preliminary lane line equation fitting effect provided in Embodiment 2 of the lane line fitting method of the present application;

[0052] Figure 5This is a schematic diagram of the module structure of the lane line fitting device according to an embodiment of the present application;

[0053] Figure 6 Schematic diagram of the device structure of the hardware operating environment involved in the lane line fitting method in the embodiment of the present application.

[0054] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0055] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0056] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0057] In the field of autonomous driving, accurate identification and fitting of lane lines are extremely critical for lateral control of vehicles. Existing systems mostly use forward-looking cameras and sensors to obtain lane line point sets and fit them through polynomial models. However, as the application scenarios expand to complex road sections, higher requirements are placed on the accuracy and real-time performance of lane line fitting. Existing technologies perform poorly when dealing with complex curves, and do not fully combine vehicle speed and curve curvature for adaptive adjustment, resulting in insufficient fitting accuracy. Therefore, it is necessary to introduce a dynamically adjusted adaptive fitting solution to improve the accuracy and stability of lane line fitting and ensure the effectiveness of the autonomous driving system under various road conditions.

[0058] The main solution of the embodiment of the present application is: first, the current signal of the window motor is obtained through the current sampling resistor to determine the initial position of the window and the motor current intensity, then the signal is filtered to extract the current ripple and volt value signal, and the cycle is counted to identify the working status of the motor, and finally the current window position is calculated based on this information.

[0059] It should be noted that the execution subject of the embodiment of the present application may be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device, a vehicle system, an automatic driving system, etc. that can realize the above functions. The automatic driving system is taken as an example to illustrate this embodiment and the following embodiments.

[0060] Based on this, the embodiment of the present application provides a lane line fitting method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the lane fitting method of the present application.

[0061] In this embodiment, the lane line fitting method includes steps S10 to S30:

[0062] Step S10, acquiring a driving image, and extracting an initial lane line point set from the driving image;

[0063] It should be noted that driving images refer to real-time images or video streams taken by the front-view camera on the vehicle. These images are used to obtain visual information of the road ahead of the vehicle. Through these images, the system can identify and analyze lane lines, traffic signs, obstacles, etc. The initial lane line point set refers to the set of points related to the lane line extracted from the driving image. These points represent the position and shape of the lane line in the image. Usually, the lane line point set is extracted from the image captured by the camera through image processing and computer vision technology. These point sets are the basis for subsequent lane line fitting and analysis, helping the system determine the specific position and shape of the lane line.

[0064] It can be understood that obtaining driving images refers to using the vehicle's front-view camera to capture real-time images or video streams of the road ahead while the vehicle is driving. These images provide visual information of the road for subsequent processing. On this basis, the system analyzes these images through image processing algorithms, identifies the characteristics of the lane lines, and extracts the initial lane line point sets from them. These point sets contain the position data of the lane lines in the image, which are obtained by edge detection, color segmentation or other visual processing technologies on the lane lines. These initial lane line point sets provide key basic data for the fitting and accurate identification of subsequent lane lines.

[0065] Step S20, obtaining a desired lane line point set from the initial lane line point set according to a preset lane line point set interval, a polynomial model, and a lane line fitting adaptive model;

[0066] The preset lane line point set interval refers to an area range pre-set by the system when the lane line point set is extracted and processed. This interval is usually determined based on the current front field of view of the vehicle (for example, 100 meters) so that an appropriate point set can be selected from the initial lane line point set for processing. The preset interval helps the system to concentrate on processing the lane line data in the road ahead and improve the accuracy and efficiency of fitting. The polynomial model is a mathematical model used to describe the shape of the lane line. It fits the lane line point set through polynomial equations (such as cubic polynomials) to establish a mathematical expression for the lane line. This model can handle lane lines of different shapes such as straight lines and curved lines, and is often used for the preliminary fitting of lane lines. The lane line fitting adaptive model refers to a dynamically adjusted model that adjusts the fitting interval of the lane line point set according to the current driving conditions of the vehicle (such as lane line curvature, vehicle speed, etc.). The model optimizes the accuracy of lane line fitting by calculating and adjusting the fitting interval in real time to adapt to different road curvatures and vehicle speed changes. The calculation formula of the model is as follows:

[0067]

[0068] Where X is the calculated expected fitting point set interval, Crv0 is the preset critical threshold curve curvature, Crv is the curve curvature calculated in the previous cycle, Veh0 is the preset critical threshold vehicle speed, Veh is the current vehicle speed, and the threshold can be obtained through actual vehicle calibration. min is the preset minimum fitting interval, X max is the preset maximum fitting interval. w1 and w2 are both greater than 0, which are the weight coefficients of curve curvature and vehicle speed respectively.

[0069] The expected lane line point set refers to the final point set obtained by screening and adjusting the initial lane line point set through the preset lane line point set interval, polynomial model and lane line fitting adaptive model. These point sets are optimized and adjusted to more accurately represent the actual position and shape of the lane line, providing a more accurate data basis for subsequent lane line fitting and analysis.

[0070] It can be understood that, first, the system selects a suitable point set in the initial lane line point set based on a preset lane line point set interval (such as a range of 100 meters ahead). This interval is to centrally process the relevant data in front of the vehicle. Next, a polynomial model is applied to perform a preliminary fitting of these point sets to establish a preliminary mathematical description of the lane line. Subsequently, combined with the lane line fitting adaptive model, the fitting interval is dynamically adjusted according to the current driving conditions of the vehicle (such as lane line curvature and vehicle speed) to optimize the selection and fitting accuracy of the point set. These adjustments ensure that the expected lane line point set can accurately reflect the actual shape of the lane line, thereby improving the accuracy and reliability of the fitting.

[0071] As an example, before the step of intercepting the expected lane line point set from the initial lane line point set according to the preset lane line point set interval, the polynomial model and the lane line fitting adaptive model, it also includes: obtaining the historical lane line point set; calculating the historical curve curvature according to the historical lane line point set; constructing the lane line fitting adaptive model according to the preset critical threshold curve curvature, the historical curve curvature, the preset critical threshold vehicle speed, the preset minimum fitting interval, the preset maximum fitting interval and the vehicle speed weight coefficient.

[0072] The historical lane line point set refers to the set of lane line points extracted from the vehicle driving images in the previous time period. These point sets record the shape and position of the lane line in a certain time period in the past, and are used to analyze and predict the shape of the current lane line. The historical curve curvature refers to the curve curvature value calculated based on the historical lane line point set, which reflects the degree of road curvature in the past time period. For example, if the lane line point set in the past 10 seconds shows the curvature change of the lane line, the calculated curvature value may be 0.03, which means that the road is moderately curved. The preset critical threshold curve curvature refers to a threshold defined in the system design, which is used to determine whether the curve curvature of the lane line reaches a level that requires special treatment. The curvature exceeding this threshold will trigger adaptive adjustment. For example, the preset critical threshold curve curvature in the design is 0.05. When the calculated actual curvature exceeds 0.05, the system will adjust the fitting interval to improve the fitting accuracy. The preset critical threshold vehicle speed refers to a vehicle speed value set in the system, which is used to determine whether the current speed of the vehicle needs to adjust the lane line fitting strategy. If the vehicle speed exceeds this threshold, the system will make corresponding adjustments. For example, the critical threshold speed set by the system is 60km / h. When the vehicle speed exceeds 60km / h, the system will recalculate and adjust the fitting interval of the lane line. The preset minimum fitting interval refers to the minimum range defined by the system during the lane line fitting process. Even when the vehicle speed is low or the curvature is small, the system will ensure that the fitting interval is not less than this minimum value. For example, if the preset minimum fitting interval is 30 meters, then in any case, the system will use a point set of at least 30 meters for fitting. The preset maximum fitting interval refers to the maximum range defined by the system during the lane line fitting process. This range is the maximum point set range allowed by the system when processing lane line data. For example, if the preset maximum fitting interval is 100 meters, then even if the vehicle speed is low or the curvature is small, the system will not use a point set of more than 100 meters for fitting. The speed weight coefficient refers to the weight parameter used to adjust the influence of vehicle speed on the lane line fitting interval. It helps the system adjust the size of the fitting interval according to the change of vehicle speed to improve the accuracy of fitting. For example, if the vehicle speed weight coefficient is set to 0.2, then the change in vehicle speed will have a certain impact on the adjustment of the fitting interval, but the larger the weight coefficient, the more significant the impact of vehicle speed on the fitting interval.

[0073] Before the optimization and interception step of the lane line point set, the system first obtains the historical lane line point set, which is the lane line data recorded over a period of time, which is used to analyze the changes in the lane line in the past. Based on these historical data, the system calculates the historical curve curvature to understand the degree of curvature of the road in the past. Then, the system uses these historical data and preset standards, such as the preset critical threshold curve curvature, the preset critical threshold vehicle speed, the preset minimum fitting interval, the preset maximum fitting interval, and the vehicle speed weight coefficient to build a lane line fitting adaptive model. This model will dynamically adjust the size of the lane line fitting interval to adapt to the current road conditions and vehicle driving status, thereby ensuring that the fitting results are more accurate. After these steps, the system will filter out the most suitable expected lane line point set from the initial lane line point set based on the preset lane line point set interval, polynomial model, and adaptive model.

[0074] Step S30, fitting the desired lane line point set according to the polynomial model to obtain the target lane line equation, and completing the lane line fitting.

[0075] It should be noted that fitting refers to processing the selected desired lane line point set through a certain mathematical model to find the best curve or equation to describe the distribution of these points. Specifically, the fitting process uses a polynomial model to perform mathematical operations on the point set to generate a curve or equation that best reflects the overall trend and shape of the point set. The goal of this process is to optimize the model parameters so that the fitting curve matches the shape of the actual lane line as accurately as possible. For example: if the desired lane line point set contains multiple lane line measurement points, a smooth curve can be obtained by fitting a cubic polynomial model. This curve can best describe the overall distribution of these points. The target lane line equation refers to the mathematical expression obtained through the fitting process, which describes the shape and position of the lane line. This equation is usually in the form of a polynomial (such as a cubic polynomial) to represent the actual shape of the lane line in front of the vehicle. The target lane line equation is the fitting result of the desired lane line point set and is the key data for lane line recognition and prediction. For example: through polynomial fitting, the target lane line equation obtained may be y=C0+C1x+C2x 2 +C3x 3 , where C0, C1, C2, C3 are polynomial coefficients, and x and y represent the longitudinal and lateral distances from the camera origin, respectively. This equation can be used to describe the actual shape of the lane line and perform lateral control.

[0076] It is understandable that during the lane fitting process, the system uses a polynomial model to mathematically fit the desired lane point set. This step first involves processing the desired point set obtained from the initial lane point set through a polynomial model. The polynomial model is a mathematical tool that adjusts the coefficients in the equation so that the fitted curve is closest to the distribution of these point sets. Specifically, the system calculates the optimal polynomial coefficients to ensure that the fitted curve can accurately describe the shape of the lane. Ultimately, the fitting process generates a target lane equation that represents the mathematical form and position of the lane. The target lane equation is the result of lane fitting, which provides an accurate lane description for subsequent vehicle control and path planning, thereby completing the entire lane fitting process.

[0077] Please refer to Figure 2 , Figure 2 The schematic diagram of the target lane line equation fitting effect provided in the first embodiment of the lane line fitting method of the present application shows the target lane line equation fitting effect. By comparing the original lane line scatter points and the fitting curve, the improvement effect of lane line fitting can be intuitively observed. In the figure, the horizontal axis represents the longitudinal distance from the camera origin, and the vertical axis represents the lateral position of the lane line under the camera's perspective. In the figure, the original lane line scatter points are presented in the form of discrete points, and the fitting curve attempts to depict the shape of the lane line. Compared with the preliminary lane line equation fitting, the target lane line equation fitting effect is significantly improved, and the fitting curve fits the original scatter points more closely, especially in areas where the curvature of the lane line changes greatly, the fitting curve can more accurately reflect the true direction of the lane line. This shows that by using the adaptive lane line fitting method, the lane line fitting interval can be dynamically adjusted in combination with the curvature of the lane where the vehicle is located and the real-time movement speed of the vehicle, which can effectively improve the accuracy of lane line fitting. In the figure, even in areas with more complex curvature changes, the fitting curve can fit the scatter points well, showing the effectiveness of this method in improving the accuracy of fitting lane lines in the preview interval of the automatic driving lateral control algorithm. This improved fitting effect is crucial for the lateral control of the autonomous driving system as it can provide more accurate lane line information, thereby improving the vehicle's autonomous driving performance in various road conditions.

[0078] As an example, the steps of fitting the expected lane line point set according to the polynomial model to obtain the target lane line equation include: constructing a design matrix and a longitudinal coordinate vector according to the expected lane line point set; solving the normal equation according to the design matrix and the longitudinal coordinate vector to obtain polynomial coefficients; substituting the polynomial coefficients into the polynomial model to obtain the target lane line equation.

[0079] The design matrix refers to the matrix used to represent the lane point set in polynomial fitting. It contains the independent variables of each point (such as lateral distance) and the combination of different orders of the polynomial model. Each row of the design matrix corresponds to a lane point, and the columns represent different polynomial terms (such as constant terms, linear terms, quadratic terms, etc.). This matrix is ​​used to input the data of the lane point set into the fitting model. The longitudinal coordinate vector refers to the vector representing the actual longitudinal coordinates of each point in the lane point set. This vector contains the actual position of each point on the lane line and is used to compare and fit with the data in the design matrix. The normal equation is a set of linear equations used to solve the coefficients in the polynomial fitting model. The normal equation obtains a coefficient matrix by operating the design matrix with the longitudinal coordinate vector. These coefficient matrices are the optimal parameters that minimize the fitting error. The normal equation usually has the following form: (X T X)β=X T y, where X is the design matrix, y is the longitudinal coordinate vector, and β is the polynomial coefficient to be determined. By solving this equation, the coefficients of the polynomial model can be obtained. The polynomial coefficients refer to the values ​​obtained by solving the normal equations, and these coefficients define the specific form of the polynomial model. Each coefficient corresponds to a term in the polynomial model (such as a constant term, a linear term, a quadratic term, etc.), and these coefficients are used to describe the mathematical equation of the lane line.

[0080] In the lane line fitting process, the design matrix and longitudinal coordinate vector are first constructed based on the data in the desired lane line point set. The design matrix contains the independent variables (such as the lateral coordinates) of each lane line point and their different power combinations, which are used to represent the terms of the polynomial model, while the longitudinal coordinate vector represents the longitudinal coordinate of each lane line point as the target value of the fitting. Then, using the design matrix and the longitudinal coordinate vector, the normal equation is constructed and solved, which is used to determine the coefficients of the polynomial model. By solving this equation, the coefficients of each polynomial term can be obtained. Finally, the obtained polynomial coefficients are substituted into the polynomial model to obtain the final target lane line equation, which describes the shape and position of the lane line and realizes lane line fitting.

[0081] This embodiment provides a lane line fitting method, which obtains a driving image and extracts an initial lane line point set from the driving image; obtains a desired lane line point set from the initial lane line point set according to a preset lane line point set interval, a polynomial model, and a lane line fitting adaptive model; fits the desired lane line point set according to the polynomial model to obtain a target lane line equation, and completes lane line fitting. First, a real-time driving image is obtained through the vehicle's front-view camera, and an initial lane line point set is extracted from it using image processing technology. These point sets reflect the actual position of the lane line on the road, provide basic data for subsequent fitting, and ensure that the vehicle can perceive the road conditions ahead in real time. Then, according to the preset lane line point set interval, a polynomial model and a lane line fitting adaptive model are used to filter out the desired lane line point set from the initial point set. The adaptive model dynamically adjusts the fitting interval, taking into account the real-time vehicle speed and curve curvature, so that the filtered point set is more in line with the actual road environment, improves the accuracy of fitting, and is particularly suitable for complex sections such as curves. Finally, the polynomial model is used to fit the expected lane line point set, and the polynomial coefficients are obtained by solving the normal equation to generate the target lane line equation. This equation accurately describes the shape of the lane line and can accurately fit the lane line, providing the vehicle with accurate lane information, ensuring the lateral control effect of the autonomous driving system, and improving driving safety and stability.

[0082] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can refer to the above introduction, and will not be repeated later. Figure 3 , Figure 3 This is a flow chart of the second embodiment of the lane line fitting method of the present application. Step S20 of the lane line fitting method includes steps S21 to S24:

[0083] Step S21, obtaining a target lane line point set from the initial lane line point set according to a preset lane line point set interval;

[0084] It should be noted that the target lane line point set refers to a subset selected from the initial lane line point set according to the preset lane line point set interval. After being selected, these point sets represent the lane line parts in front of the vehicle that need to be focused on, usually located in the key area of ​​the vehicle's driving route. By selecting these point sets, the system can centrally process and fit the lane line data related to the vehicle's current driving, so as to perform further fitting calculations and lane line predictions.

[0085] It can be understood that first, based on the preset lane line point set interval, the system selects the point set that meets the interval from the initial lane line point set. This preset interval is usually set according to a certain distance range in front of the vehicle (such as 100 meters in front) or a specific road section to ensure that the selected point set can accurately reflect the lane line information on the vehicle's current driving path. By intercepting the lane line point set that meets this interval, the system filters out remote or unnecessary data that is not related to the current driving, thereby obtaining the target lane line point set. These target point sets will serve as the basis for subsequent lane line fitting and vehicle control, providing more accurate lane line data for the autonomous driving system.

[0086] Step S22, fitting the target lane line point set according to a polynomial model to obtain a preliminary lane line equation;

[0087] It should be noted that the preliminary lane line equation refers to a preliminary mathematical expression generated based on the polynomial model after fitting the target lane line point set. This equation describes the basic shape and trend of the lane line by fitting the data in the target lane line point set. The preliminary lane line equation is the first step of lane line fitting. Although it can reflect the shape of the lane line well, it may need further adjustment or optimization because it does not combine other dynamic factors (such as vehicle speed, curvature, etc.).

[0088] It can be understood that, first, based on the data in the target lane line point set, a polynomial model is used to fit these point sets. The polynomial model is a mathematical tool that generates a curve that is closest to the distribution of the point set by adjusting multiple parameters (such as constant terms, linear terms, quadratic terms, etc.). During the fitting process, the system brings the horizontal and vertical coordinates of each point in the target lane line point set into the model, and calculates the optimal polynomial coefficients to form a mathematical equation, namely the preliminary lane line equation. This equation can roughly describe the shape and position of the lane line. Although this preliminary equation can already reflect the basic trend of the lane line, it is calculated based on the current target point set, and may need to be further adjusted in combination with other factors to improve the fitting accuracy and actual application effect of the lane line.

[0089] Please refer to Figure 4 , Figure 4A schematic diagram of the preliminary lane line equation fitting effect provided for the lane line fitting method embodiment 2 of the present application is provided. The figure shows the preliminary lane line equation fitting effect, in which it can be clearly seen that there is a large deviation between the fitting curve and the original lane line scatter points. The horizontal axis in the figure represents the longitudinal distance from the camera origin, and the vertical axis represents the lateral position of the lane line under the camera's view. In the figure, the original lane line scatter points are presented in the form of discrete points, while the fitting curve attempts to depict the shape of the lane line. However, it can be observed that in multiple areas, the fitting curve fails to fit the original scatter points closely, especially in areas where the curvature of the lane line changes greatly, the deviation between the fitting curve and the scatter points is particularly obvious. This shows that the preliminary lane line equation is not sufficiently accurate in fitting when dealing with complex or curvy lane lines, and cannot accurately reflect the true direction of the lane line. This lack of fitting effect may have an adverse effect on the lateral control algorithm of the autonomous driving system, because it relies on accurate lane line information to make accurate driving decisions. Therefore, it is necessary to further optimize the lane line fitting algorithm to improve the fitting accuracy and autonomous driving performance under various road conditions.

[0090] Step S23, obtaining an expected fitting point set interval according to the preliminary lane line equation, the polynomial model, the target lane line point set and the lane line fitting adaptive model;

[0091] It should be noted that the expected fitting point set interval refers to a dynamically adjusted point set range determined based on the preliminary lane line equation, polynomial model, target lane line point set and lane line fitting adaptive model. This interval represents the point set range for fitting that the system ultimately determines in the process of lane line fitting in order to improve fitting accuracy and adapt to actual road conditions.

[0092] It is understandable that the system first uses the preliminary lane line equation and polynomial model to perform a preliminary fit on the target lane line point set to generate the basic mathematical expression of the lane line. Then, combined with the lane line fitting adaptive model, the model will be optimized and adjusted according to real-time road characteristics (such as the curvature of the lane line, the current speed of the vehicle and other dynamic factors). Specifically, the adaptive model analyzes the results of the preliminary fitting, determines which lane line point sets are more important, and adjusts the fitting range according to these parameters. Through this adjustment, the system finally determines an expected fitting point set interval, that is, the part of the lane line that needs to be fitted. This interval is dynamically adjusted to ensure that the fitting focus is concentrated on the most important area for vehicle driving, thereby improving the accuracy and adaptability of lane line fitting, especially in complex road conditions.

[0093] As an example, the step of obtaining the expected fitting point set interval according to the preliminary lane line equation, the polynomial model, the target lane line point set and the lane line fitting adaptive model includes: calculating the curve curvature of the fitted lane line according to the preliminary lane line equation; when the curve curvature is greater than a preset curvature threshold, obtaining the expected fitting point set interval according to the polynomial model, the curve curvature and the lane line fitting adaptive model; when the curve curvature is less than or equal to the preset curvature threshold, using the preset lane line point set interval as the expected fitting point set interval.

[0094] The fitted lane line refers to the lane line obtained by fitting the target lane line point set through the preliminary lane line equation and the polynomial model. This fitted lane line is a mathematical description of the shape of the lane line, reflecting the overall trend and shape of the current lane line. Curve curvature refers to the degree of curvature of the lane line, which is usually used to describe the curvature of the lane line. The larger the curvature, the sharper the curve, and the smaller the curvature, the gentler the curve. In the fitting process, calculating the curve curvature of the lane line can help the system determine the degree of curvature of the road so as to dynamically adjust the fitting interval. The preset curvature threshold refers to a critical value pre-set in the system to determine the degree of curvature of the lane line, such as 0.05. When the actual calculated curve curvature is greater than this threshold, it means that the curve is sharp and the fitting interval needs to be adjusted dynamically; when the curvature is less than or equal to the threshold, it means that the curve is gentle and a fixed fitting interval can be used. The expected fitting point set interval refers to the point set range for lane line fitting determined by the system based on the actual lane line curvature and dynamic model adjustment. This range determines the lane line area that the system will focus on. If the curvature of the curve is large, the interval is dynamically narrowed according to the polynomial model and the adaptive model to improve the fitting accuracy; if the curvature is small, the preset fixed interval is used for fitting.

[0095] The system first calculates the curve curvature of the fitted lane line through the preliminary lane line equation to determine the actual curvature of the lane line. If the calculated curve curvature is greater than the curvature threshold preset by the system (that is, it means that the curve is sharp), the system will dynamically adjust the fitting range based on the polynomial model, curve curvature and lane line fitting adaptive model to determine the expected fitting point set interval to ensure that the fitting accuracy is high enough. If the curve curvature is less than or equal to the preset curvature threshold (that is, the curve is relatively gentle), the system directly uses the fixed preset lane line point set interval without dynamic adjustment. This approach ensures that the lane line fitting process can be adaptively adjusted under different curve conditions, which not only improves the fitting accuracy in curves, but also improves the calculation efficiency on flat sections.

[0096] As an example, the step of obtaining an expected fitting point set interval according to the polynomial model, the curve curvature and the lane line fitting adaptive model includes: obtaining the vehicle speed and the camera origin position, and inputting the curve curvature and the vehicle speed into the lane line fitting adaptive model to obtain an initial expected interval; calculating the fitting root mean square error according to the initial expected interval, the polynomial model and the camera origin position; when the fitting root mean square error is greater than a preset error, adjusting the vehicle speed weight coefficient of the curve curvature and the lane line fitting adaptive model, and returning to the step of inputting the curve curvature and the vehicle speed into the lane line fitting adaptive model; when the fitting root mean square error is less than or equal to the preset error, using the initial expected interval as the expected fitting point set interval.

[0097] Vehicle speed refers to the speed of the vehicle at the current moment. Vehicle speed is very important in the lane line fitting process, because a higher speed means that the vehicle needs to process the road ahead information faster, which affects the interval size and fitting accuracy of the lane line fitting. The camera origin position refers to the reference position of the front-view camera installed on the vehicle, usually the origin in the vehicle coordinate system. This position is used to calculate the coordinate position of the lane line relative to the vehicle and affects the processing of road data during lane line fitting. The initial expected interval refers to the interval of the fitting point set calculated for the first time in the lane line fitting adaptive model based on factors such as the current vehicle speed and the curvature of the curve. This interval provides a preliminary fitting range to determine which lane line point sets should be given priority consideration. The root mean square error (RMSE) of the fit refers to the error index calculated during the lane line fitting process, which is used to measure the difference between the fitting curve and the actual lane line point set. The smaller the root mean square error, the higher the fitting accuracy and the closer the fitting curve is to the actual lane line. The formula for calculating the root mean square error of the fit is as follows:

[0098]

[0099] where Y RMSE is the root mean square error of the horizontal fitting error, N is the number of scattered points of the fitting, i is the number of points, Y moeel,i The horizontal distance between the point calculated by the fitting model and the origin of the camera, Y obs,i It is the horizontal distance between the collected original scattered points and the camera origin.

[0100] The preset error refers to an error threshold pre-set in the system to determine whether the fitting result is accurate enough. When the RMS error of the fitting is less than or equal to the preset error, it means that the fitting result is within the acceptable error range and no further adjustment is required. The speed weight coefficient refers to the parameter used to adjust the influence of the speed on the fitting interval in the lane line fitting adaptive model. By increasing or decreasing the speed weight coefficient, the system can dynamically adjust the influence of the speed on the fitting result to obtain a more accurate fitting result. A larger weight coefficient means that the speed has a greater influence on the fitting interval.

[0101] The system first obtains the vehicle speed and the camera origin position, and then inputs the vehicle speed and curve curvature into the lane line fitting adaptive model to calculate a preliminary expected fitting point set interval, namely the initial expected interval. Then, the system calculates the fitting root mean square error (RMSE) based on this initial expected interval, the polynomial model and the camera origin position, which measures the difference between the fitted lane line and the actual lane line point set. If the calculated RMSE is greater than the preset error, the system will adjust the influence weight of the curve curvature and vehicle speed on the fitting (vehicle speed weight coefficient), and re-input the adjusted vehicle speed and curve curvature into the model to calculate the expected interval again. If the RMSE is less than or equal to the preset error, it means that the fitting accuracy is high enough. At this time, the system uses the initial expected interval as the final expected fitting point set interval, that is, the optimal point set range for lane line fitting. This process ensures that the accuracy of lane line fitting is high enough by continuously adjusting the weights and intervals, especially in complex road conditions.

[0102] As an example, the step of calculating the fitting root mean square error according to the initial expected interval, the polynomial model and the camera origin position includes: intercepting the original lane line point set from the initial lane line point set according to the initial expected interval; fitting the original lane line point set according to the polynomial model to obtain a fitting lane line equation; calculating the fitting lane line point set according to the fitting lane line equation; calculating the fitting root mean square error according to the fitting lane line point set, the original lane line point set and the camera origin position.

[0103] The original lane line point set refers to a part of lane line data points selected from the initial lane line point set within the initial expected interval. These point sets reflect the actual position and shape of the lane line within a preset range in front of the vehicle and are used for fitting calculations. The fitted lane line equation refers to the mathematical expression obtained by fitting the original lane line point set through a polynomial model. This equation is used to describe the shape and position of the lane line and is the result of the fitting calculation. Usually, this equation is in the form of a polynomial, such as a quadratic or cubic polynomial, which can accurately describe the curvature and change of the lane line. The fitted lane line point set refers to the point set calculated according to the fitted lane line equation. By substituting the lateral coordinates of the original lane line point set into the fitted lane line equation, the system can generate the fitted point set, which represents the lane line data points corresponding to the fitted curve at each position.

[0104] First, based on the initial expected interval, the part that needs to be processed currently is screened out from the initial lane line point set, that is, the original lane line point set. Then, the system uses a polynomial model to fit these original lane line point sets to generate a fitted lane line equation that can describe the shape of the lane line. Then, based on the fitting equation, the system calculates the fitted lane line point set corresponding to each point, that is, the lane line data obtained by fitting. Finally, the system compares the fitted lane line point set with the original lane line point set, and calculates the fitting root mean square error (RMSE) combined with the origin position of the vehicle camera. This is an error indicator that reflects the fitting accuracy. A smaller RMSE value indicates that the fitting curve matches the actual lane line point set better, and a larger error indicates that the fitting needs further adjustment.

[0105] Step S24, obtaining an expected lane line point set from the initial lane line point set according to the expected fitting point set interval.

[0106] It can be understood that according to the expected fitting point set interval calculated previously, lane line data points within a specific range are screened out from the entire initial lane line point set, which is called the expected lane line point set. The expected fitting point set interval is dynamically adjusted according to the current driving state of the vehicle (such as speed and curve curvature) and the preset adaptive model. Through this interval, the system can accurately intercept the lane line point set that is closely related to the vehicle's driving, remove unnecessary or remote lane line data, and thus provide more accurate and effective input data for subsequent fitting.

[0107] In this embodiment, the target lane line point set is intercepted from the initial lane line point set according to the preset lane line point set interval; the target lane line point set is fitted according to the polynomial model to obtain the preliminary lane line equation; the expected fitting point set interval is obtained according to the preliminary lane line equation, the polynomial model, the target lane line point set and the lane line fitting adaptive model; the expected lane line point set is intercepted from the initial lane line point set according to the expected fitting point set interval. In the lane line fitting process, the system first selects the target lane line point set from the initial lane line point set according to the preset lane line point set interval. This process improves the calculation efficiency and fitting accuracy by focusing on the data points related to the vehicle driving path. Then, the system fits the target lane line point set using the polynomial model to generate the preliminary lane line equation, which provides a mathematical description for the basic shape and position of the lane line. Subsequently, the system combines the preliminary lane line equation, the fitting model and the adaptive model, calculates the expected fitting point set interval according to the real-time vehicle speed and road curvature, and dynamically adjusts the fitting range to adapt to complex road conditions, thereby improving the accuracy of fitting. Finally, the system further selects the desired lane line point set from the initial lane line point set based on the calculated expected fitting point set interval, ensuring that the data points used for fitting are the most relevant and accurate. These steps gradually optimize data processing, ensuring that the system can accurately fit lane lines, and improving the path planning and control accuracy of the autonomous driving system under different road conditions.

[0108] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the lane line fitting method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0109] This application also provides a lane line fitting device, please refer to Figure 5 , the lane line fitting device comprises:

[0110] A data acquisition module 10 is used to acquire a driving image and extract an initial lane line point set from the driving image;

[0111] A data processing module 20, configured to obtain a desired lane line point set from the initial lane line point set according to a preset lane line point set interval, a polynomial model, and a lane line fitting adaptive model;

[0112] The fitting module 30 is used to fit the desired lane line point set according to the polynomial model to obtain the target lane line equation and complete the lane line fitting.

[0113] In one embodiment, the data processing module 20 is further used to obtain a target lane line point set from the initial lane line point set according to a preset lane line point set interval; fit the target lane line point set according to a polynomial model to obtain a preliminary lane line equation; obtain an expected fitting point set interval according to the preliminary lane line equation, the polynomial model, the target lane line point set and the lane line fitting adaptive model; and obtain an expected lane line point set from the initial lane line point set according to the expected fitting point set interval.

[0114] In one embodiment, the data processing module 20 is also used to calculate the curve curvature of the fitted lane line based on the preliminary lane line equation; when the curve curvature is greater than a preset curvature threshold, the expected fitting point set interval is obtained according to the polynomial model, the curve curvature and the lane line fitting adaptive model; when the curve curvature is less than or equal to the preset curvature threshold, the preset lane line point set interval is used as the expected fitting point set interval.

[0115] In one embodiment, the data processing module 20 is also used to obtain the vehicle speed and the camera origin position, and input the curve curvature and the vehicle speed into the lane line fitting adaptive model to obtain an initial expected interval; calculate the fitting root mean square error according to the initial expected interval, the polynomial model and the camera origin position; when the fitting root mean square error is greater than a preset error, adjust the curve curvature and the vehicle speed weight coefficient of the lane line fitting adaptive model, and return to the step of inputting the curve curvature and the vehicle speed into the lane line fitting adaptive model; when the fitting root mean square error is less than or equal to the preset error, use the initial expected interval as the expected fitting point set interval.

[0116] In one embodiment, the data processing module 20 is further used to obtain an original lane line point set from the initial lane line point set according to the initial expected interval; fit the original lane line point set according to the polynomial model to obtain a fitted lane line equation; calculate the fitted lane line point set according to the fitted lane line equation; and calculate the fitting root mean square error according to the fitted lane line point set, the original lane line point set and the camera origin position.

[0117] In one embodiment, the data processing module 20 is also used to obtain a historical lane line point set; calculate a historical curve curvature based on the historical lane line point set; and construct a lane line fitting adaptive model based on a preset critical threshold curve curvature, the historical curve curvature, a preset critical threshold vehicle speed, a preset minimum fitting interval, a preset maximum fitting interval, and a vehicle speed weight coefficient.

[0118] In one embodiment, the fitting module 30 is further used to construct a design matrix and a longitudinal coordinate vector according to the desired lane line point set; solve the normal equation according to the design matrix and the longitudinal coordinate vector to obtain polynomial coefficients; substitute the polynomial coefficients into the polynomial model to obtain the target lane line equation.

[0119] The lane line fitting device provided by the present application adopts the lane line fitting method in the above embodiment, which can solve the technical problem of how to accurately fit the lane line. Compared with the prior art, the beneficial effects of the lane line fitting device provided by the present application are the same as the beneficial effects of the lane line fitting method provided by the above embodiment, and other technical features in the lane line fitting device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0120] The present application provides a lane line fitting device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the lane line fitting method in the above-mentioned embodiment 1.

[0121] Reference below Figure 6 , which shows a schematic diagram of the structure of a lane line fitting device suitable for implementing the embodiment of the present application. The lane line fitting device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The lane fitting device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0122] like Figure 6As shown, the lane fitting device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the lane fitting device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the lane fitting device to communicate with other devices wirelessly or by wire to exchange data. Although the lane fitting device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0123] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0124] The lane line fitting device provided by the present application adopts the lane line fitting method in the above embodiment, which can solve the technical problem of how to accurately fit the lane line. Compared with the prior art, the beneficial effects of the lane line fitting device provided by the present application are the same as the beneficial effects of the lane line fitting method provided by the above embodiment, and the other technical features in the lane line fitting device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0125] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0126] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0127] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the lane line fitting method in the above-mentioned embodiment.

[0128] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0129] The computer-readable storage medium may be included in the lane line fitting device; or may exist independently without being assembled into the lane line fitting device.

[0130] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the lane line fitting device, the lane line fitting device: obtains a driving image and extracts an initial lane line point set from the driving image; intercepts a desired lane line point set from the initial lane line point set according to a preset lane line point set interval, a polynomial model and a lane line fitting adaptive model; fits the desired lane line point set according to the polynomial model to obtain a target lane line equation, and completes lane line fitting.

[0131] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0132] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0133] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0134] The readable storage medium provided in this application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above lane line fitting method, and can solve the technical problem of how to accurately fit the lane line. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the lane line fitting method provided in the above embodiment, and will not be repeated here.

[0135] The present application also provides a computer program product, including a computer program, which implements the steps of the lane line fitting method as described above when executed by a processor.

[0136] The computer program product provided by the present application can solve the technical problem of how to accurately fit the lane line. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the lane line fitting method provided by the above embodiment, which will not be repeated here.

[0137] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A lane line fitting method, characterized in that: The method comprises: Acquire a driving image, and extract an initial lane line point set from the driving image; Obtaining a desired lane line point set from the initial lane line point set according to a preset lane line point set interval, a polynomial model, and a lane line fitting adaptive model; The desired lane line point set is fitted according to the polynomial model to obtain a target lane line equation, thereby completing lane line fitting.

2. The method according to claim 1, characterized in that The step of obtaining a desired lane line point set from the initial lane line point set according to the preset lane line point set interval, the polynomial model and the lane line fitting adaptive model comprises: Obtaining a target lane line point set from the initial lane line point set according to a preset lane line point set interval; Fitting the target lane line point set according to a polynomial model to obtain a preliminary lane line equation; Obtaining an expected fitting point set interval according to the preliminary lane line equation, the polynomial model, the target lane line point set, and the lane line fitting adaptive model; The expected lane line point set is obtained by intercepting the initial lane line point set according to the expected fitting point set interval.

3. The method according to claim 2, characterized in that The step of obtaining the expected fitting point set interval according to the preliminary lane line equation, the polynomial model, the target lane line point set and the lane line fitting adaptive model comprises: Calculate the curvature of the curve of the fitted lane line according to the preliminary lane line equation; When the curve curvature is greater than a preset curvature threshold, obtaining an expected fitting point set interval according to the polynomial model, the curve curvature and the lane line fitting adaptive model; When the curve curvature is less than or equal to the preset curvature threshold, the preset lane line point set interval is used as the expected fitting point set interval.

4. The method according to claim 3, characterized in that The step of obtaining the expected fitting point set interval according to the polynomial model, the curve curvature and the lane line fitting adaptive model comprises: Obtaining the vehicle speed and the camera origin position, and inputting the curve curvature and the vehicle speed into the lane line fitting adaptive model to obtain an initial expected interval; Calculate the fitting root mean square error according to the initial expected interval, the polynomial model and the camera origin position; When the fitting root mean square error is greater than a preset error, adjusting the vehicle speed weight coefficient of the curve curvature and the lane line fitting adaptive model, and returning to the step of inputting the curve curvature and the vehicle speed into the lane line fitting adaptive model; When the fitting root mean square error is less than or equal to the preset error, the initial expected interval is used as the expected fitting point set interval.

5. The method according to claim 4, characterized in that The step of calculating the fitting root mean square error according to the initial expected interval, the polynomial model and the camera origin position comprises: Obtaining an original lane line point set from the initial lane line point set according to the initial expected interval; Fitting the original lane line point set according to the polynomial model to obtain a fitted lane line equation; Calculate a fitted lane line point set according to the fitted lane line equation; A fitting root mean square error is calculated based on the fitting lane line point set, the original lane line point set and the camera origin position.

6. The method according to claim 1, characterized in that Before the step of obtaining the desired lane line point set from the initial lane line point set according to the preset lane line point set interval, the polynomial model and the lane line fitting adaptive model, the method further includes: Get the historical lane line point set; Calculating the historical curve curvature according to the historical lane line point set; A lane line fitting adaptive model is constructed according to a preset critical threshold curve curvature, the historical curve curvature, a preset critical threshold vehicle speed, a preset minimum fitting interval, a preset maximum fitting interval and a vehicle speed weight coefficient.

7. The method according to any one of claims 1 to 6, characterized in that The step of fitting the desired lane line point set according to the polynomial model to obtain the target lane line equation comprises: Constructing a design matrix and a longitudinal coordinate vector according to the desired lane line point set; Solving normal equations according to the design matrix and the longitudinal coordinate vector to obtain polynomial coefficients; Substitute the polynomial coefficients into the polynomial model to obtain the target lane line equation.

8. A lane line fitting device, characterized in that: The device comprises: A data acquisition module, used to acquire a driving image and extract an initial lane line point set from the driving image; A data processing module, used for intercepting a desired lane line point set from the initial lane line point set according to a preset lane line point set interval, a polynomial model and a lane line fitting adaptive model; The fitting module is used to fit the desired lane line point set according to the polynomial model to obtain the target lane line equation and complete the lane line fitting.

9. A lane line fitting device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the lane line fitting method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the lane line fitting method according to any one of claims 1 to 7 are implemented.

Images