Lane line position detection method and device, electronic equipment and storage medium

Abstract

The invention discloses a lane line position detection method and device, electronic equipment and a storage medium, and relates to the field of automobile driving and road recognizing.The method comprises the steps that a region of interest of a current vehicle is obtained according to vehicle data, and detection data of the region of interest is obtained through a detection radar; obtaining a candidate guardrail point set according to the detection data, and obtaining an associated guardrail point set based on the guardrail point features according to the candidate guardrail point set; and obtaining a guardrail position based on a polynomial fitting mode according to the associated guardrail point set so as to obtain a lane line position corresponding to the current vehicle according to the guardrail position. According to the technical scheme, real-time detection of the lane line position is achieved, dynamic obtaining of the lane line position is achieved on the basis of the lane line position obtained through calculation of the real-time guardrail position, the accuracy of the lane line detection result is greatly improved, and vehicle driving safety is ensured.

Description

Technical Field

[0001] This invention relates to the field of automobile driving and road recognition, and in particular to a lane line position detection method, device, electronic device and storage medium. Background Technology

[0002] With the rapid development of the automotive industry, intelligent autonomous driving technology is becoming a hot research topic in the automotive field. Among them, lane lines are important markers that constrain vehicles during driving. Controlling vehicles to drive within lane lines based on the detected lane line positions is a necessary prerequisite for ensuring the safety of autonomous driving.

[0003] In existing technologies, lane line position detection usually relies on cameras mounted on the vehicle body for recognition. However, when lane lines cannot be effectively identified in severe weather, such as heavy rain causing cameras to be unable to clearly identify lane lines, or snow burying lane lines after a blizzard, a calibration value is usually pre-configured to predict the location of the lane lines.

[0004] However, based on the pre-configured calibration data, the location of the lane lines cannot be accurately calculated. This calculation method makes it impossible to guarantee the accuracy and safety of the calculation results. Therefore, relying on cameras to perceive lane lines will lead to a decrease in the performance and stability of the autonomous driving system. Summary of the Invention

[0005] This invention provides a lane line position detection method, apparatus, electronic device, and storage medium to solve the problem of low accuracy in lane line position detection results.

[0006] According to another aspect of the present invention, a lane line position detection method is provided, comprising:

[0007] The region of interest for the current vehicle is obtained based on the vehicle data, and the detection data of the region of interest is obtained through the detection radar;

[0008] Based on the detection data, a set of candidate guardrail points is obtained, and based on the guardrail point features, a set of associated guardrail points is obtained from the set of candidate guardrail points.

[0009] Based on the associated guardrail point set, the guardrail position is obtained using a multinomial fitting method, and the lane line position corresponding to the current vehicle is obtained based on the guardrail position.

[0010] The step of obtaining the region of interest (ROI) of the current vehicle based on the whole vehicle data includes: obtaining the vehicle's turning radius based on the whole vehicle data, and obtaining the vehicle's driving curvature based on the vehicle's turning radius, so as to obtain the ROI of the current vehicle based on the vehicle's driving curvature; wherein, if the vehicle's driving curvature is less than or equal to a preset curvature threshold, the left front region and the right front region are taken as ROI; if the vehicle's driving curvature is greater than the preset curvature threshold and the vehicle's turning direction is left, the left front region and the left rear region are taken as ROI; if the vehicle's driving curvature is greater than the preset curvature threshold and the vehicle's turning direction is right, the right front region and the right rear region are taken as ROI.

[0011] The vehicle data includes vehicle speed, steering wheel angle, gear ratio, front wheel track width, and yaw rate. Obtaining the vehicle turning radius based on the vehicle data includes: if the vehicle speed is determined to be less than a preset speed threshold, obtaining the vehicle turning radius based on the steering wheel angle, the gear ratio, and the front wheel track width; if the vehicle speed is determined to be greater than or equal to the preset speed threshold, obtaining the vehicle turning radius based on the vehicle speed and the yaw rate.

[0012] The step of obtaining a candidate guardrail point set based on the detection data includes: obtaining a candidate guardrail point set based on the detection data using a radar cross section threshold and a distance change rate threshold.

[0013] The step of obtaining a set of associated guardrail points based on the guardrail point features of the candidate guardrail point set includes: obtaining the Mahalanobis distance between different candidate guardrail points based on the position, distance change rate, and angle of each candidate guardrail point; and obtaining the set of associated guardrail points based on the Mahalanobis distance between each candidate guardrail point.

[0014] After obtaining the guardrail position based on the associated guardrail point set using a multinomial fitting method, and then obtaining the lane line position corresponding to the current vehicle based on the guardrail position, the method further includes: if a new candidate guardrail point is detected in the candidate guardrail point set, determining whether to add the new candidate guardrail point to the associated guardrail point set based on the Mahalanobis distance between the new candidate guardrail point and each associated guardrail point; if it is determined that the new candidate guardrail point has been added to the associated guardrail point set, updating the guardrail position based on the associated guardrail point set using an incremental fitting method, and then obtaining the lane line position corresponding to the current vehicle based on the updated guardrail position.

[0015] According to another aspect of the present invention, a lane line position detection device is provided, comprising:

[0016] The detection data acquisition module is used to acquire the region of interest of the current vehicle based on the whole vehicle data, and to acquire the detection data of the region of interest through the detection radar;

[0017] The guardrail point set acquisition module is used to acquire candidate guardrail point sets based on the detection data, and to acquire associated guardrail point sets based on guardrail point features according to the candidate guardrail point sets.

[0018] The guardrail position acquisition module is used to obtain the guardrail position based on the associated guardrail point set using a multinomial fitting method, so as to obtain the lane line position corresponding to the current vehicle based on the guardrail position.

[0019] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the lane line position detection method according to any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the lane line position detection method according to any embodiment of the present invention.

[0021] According to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the lane line position detection method described in any embodiment of the present invention.

[0022] The technical solution of this invention first obtains the region of interest (ROI) of the current vehicle based on vehicle data, and then acquires detection data of the ROI using a detection radar. Next, it obtains a candidate guardrail point set based on the detection data, and then filters and obtains a set of associated guardrail points. Finally, it obtains the guardrail position based on the guardrail point set using a multinomial fitting method, and then uses the guardrail position to obtain the lane line position corresponding to the current vehicle. This not only achieves real-time detection of lane line positions, but also enables dynamic acquisition of lane line positions based on the real-time guardrail positions. Compared to pre-configured scalar values, this significantly improves the accuracy of lane line detection results, ensuring vehicle driving safety, especially in autonomous driving scenarios, and thus enhancing the performance and stability of the autonomous driving system.

[0023] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of a lane line position detection method provided in Embodiment 1 of the present invention;

[0026] Figure 2 This is a flowchart of another lane line position detection method provided in Embodiment 2 of the present invention;

[0027] Figure 3 This is a flowchart of another lane line position detection method provided in Embodiment 3 of the present invention;

[0028] Figure 4 This is a schematic diagram of the structure of a lane line position detection device according to Embodiment 4 of the present invention;

[0029] Figure 5 This is a schematic diagram of the structure of an electronic device that implements the lane line position detection method of this invention. Detailed Implementation

[0030] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0031] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0032] Example 1

[0033] Figure 1 This is a flowchart of a lane line position detection method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where the lane line position is obtained by measuring the position of guardrails on both sides of the road. The method can be executed by a lane line position detection device, which can be implemented in hardware and / or software. This lane line position detection device can be configured in an electronic device (e.g., an in-vehicle terminal device). Figure 1 As shown, the method includes:

[0034] S101. Obtain the region of interest of the current vehicle based on the whole vehicle data, and obtain the detection data of the region of interest through the detection radar.

[0035] Vehicle data includes vehicle driving data and vehicle static data. Vehicle driving data can include vehicle speed, yaw rate, and steering wheel angle. Vehicle static data can include front wheel track and gear ratio. Yaw rate refers to the angular velocity of the vehicle around its vertical axis, reflecting the dynamic response of the vehicle when turning. Gear ratio is the ratio of the rotational speed of the input shaft to the output shaft of the transmission, which determines how engine power is transmitted to the wheels. The front wheel angle can be obtained based on the ratio of steering wheel angle to gear ratio.

[0036] Detection radar is a technical device installed on the outside of a vehicle, such as in the front area (both sides and center of the front bumper), the rear area (both sides of the rear bumper), and the lateral areas (from the vehicle's perspective and the B-pillars of the doors). It is used to acquire information about the surrounding environment by emitting detection waves. It calculates information such as the distance, speed, and orientation of obstacles by emitting signals and receiving reflected waves. Detection radar can include various types such as ultrasonic radar, millimeter-wave radar, and lidar. The detection data acquired by detection radar can specifically include point information and trajectory information.

[0037] Point information refers to the target reflection points detected in a single scan of a detection radar. It reflects the instantaneous measurement values ​​of the detected target and can include range, azimuth, rate of change of range, and radar cross section (RCS). Among them, range refers to the straight-line distance between the detected target and the detection radar; azimuth refers to the angle between the detected target and the detection radar in the horizontal direction; the rate of change of range is the radial velocity of the detected target obtained based on the Doppler effect; and the radar cross section is a measure of the detected target's ability to reflect electromagnetic waves.

[0038] The trajectory information is the target motion trajectory formed by multi-frame point correlation filtering. It reflects the continuous state of the detected target and can include average distance, angle, average distance change rate, and average radar cross section. Among them, the average distance, average distance change rate, and average radar cross section are all average values ​​calculated based on multi-frame point information to reduce single-frame measurement noise and improve the detection stability of low-speed targets. The angle includes the horizontal angle and pitch angle of the detected target.

[0039] The vehicle's trajectory can be determined based on vehicle data (e.g., steering wheel angle). If the vehicle is turning left, the left-side region (e.g., the front left and rear left regions) can be designated as the region of interest. If the vehicle is turning right, the right-side region (e.g., the front right and rear right regions) can be designated as the region of interest. If the vehicle is traveling in a straight line, both sides (e.g., the front left, front right, rear left, and rear right regions) can be designated as the regions of interest. After determining the region of interest, a detection radar on the same side as the region of interest is selected to acquire its detection data.

[0040] S102. Obtain a candidate guardrail point set based on the detection data, and obtain an associated guardrail point set based on the guardrail point features of the candidate guardrail point set.

[0041] After acquiring the detection data from the radar, the data can be preprocessed. First, the detection data is denoised. For example, based on the distance change rate of continuous point information, fluctuating points with abrupt changes in value are deleted. Based on the radar cross section threshold of continuous point information, isolated points with abrupt changes in value are deleted. Then, the polar coordinates (e.g., distance and azimuth) are converted to rectangular coordinates in the vehicle coordinate system. The vehicle coordinate system can be based on the rear axle center of the vehicle as the origin, the horizontal axis as the forward direction, and the vertical axis as the turning direction (i.e., left or right). Based on this, the preprocessed detection points can be used as a candidate guardrail point set.

[0042] Because guardrails are stationary, moving points (e.g., vehicles and pedestrians) in the detection data can be removed. At the same time, guardrails themselves have the characteristic of continuous appearance, existing in a large area (e.g., large volume and area) and having a certain height above the ground. Based on this, other types of stationary points (e.g., curbs with low height above the ground; small volume and area of ​​traffic barriers) can be removed. The candidate guardrail point set after removing moving points and other types of stationary points is then used as the associated guardrail point set.

[0043] Optionally, in this embodiment of the invention, obtaining a candidate guardrail point set based on the detection data includes: obtaining a candidate guardrail point set based on the detection data using a radar cross section threshold and a distance change rate threshold. The radar cross section describes the intensity of radar electromagnetic wave reflection from a target object. Since all guardrails are made of the same material, they correspond to the same radar cross section, which is distinct from the radar cross sections of other objects. The distance change rate describes the rate of change of the distance between the target object and the detection radar over time. Since the position of the guardrail relative to the lane is fixed, the distance change rate of the guardrail points is close to 0.

[0044] Therefore, points that are stable over a long period of time, spatially concentrated, and have a distance change rate less than or equal to the distance change rate threshold, while also meeting the radar cross section (RCS) threshold requirements, can be selected as candidate guardrail points. For example, detection points with an RCS value between -10 and 20 dBsm and an absolute distance change rate less than 0.1 meters per second can be selected as candidate guardrail points. By obtaining the candidate guardrail point set through the RCS threshold and the distance change rate threshold, accurate screening of candidate guardrail points is achieved, avoiding the problem of poor guardrail location efficiency caused by an excessive number of candidate guardrail points.

[0045] S103. Obtain the guardrail position based on the associated guardrail point set using a multinomial fitting method, and obtain the lane line position corresponding to the current vehicle based on the guardrail position.

[0046] After obtaining the set of associated guardrail points, substitute each associated guardrail point into the following polynomial equation:

[0047] ;

[0048] in, Represents the x-coordinate of the associated guardrail point; Represents the ordinate of the associated guardrail point; The unknown coefficients represent the polynomials; these coefficients can be solved using the least squares method, from which the guardrail position equation representing the guardrail's location can be obtained; specifically, It reflects the rate of change of the trajectory curvature; This reflects the overall tilting trend of the guardrail; This reflects the lateral offset of the guardrail.

[0049] Since the lateral distance between the guardrail and the lane lines is fixed—for example, on highways, the lateral distance between the guardrail and the lane lines is 0.5 meters to 1.5 meters—the specific lateral distance can be obtained through map data or offline calibration data. Based on the location of the guardrail and the aforementioned lateral distance The current lane position of the vehicle can be obtained as follows:

[0050] ;

[0051] Specifically, the aforementioned lateral distance It can also be dynamically corrected by yaw rate, for example, by the lateral distance when the vehicle is tilted. Side roll compensation needs to be added to ensure the accuracy of lane line position detection.

[0052] The technical solution of this invention first obtains the region of interest (ROI) of the current vehicle based on vehicle data, and then acquires detection data of the ROI using a detection radar. Next, it obtains a candidate guardrail point set based on the detection data, and then filters and obtains a set of associated guardrail points. Finally, it obtains the guardrail position based on the guardrail point set using a multinomial fitting method, and then uses the guardrail position to obtain the lane line position corresponding to the current vehicle. This not only achieves real-time detection of lane line positions, but also enables dynamic acquisition of lane line positions based on the real-time guardrail positions. Compared to pre-configured scalar values, this significantly improves the accuracy of lane line detection results, ensuring vehicle driving safety, especially in autonomous driving scenarios, and thus enhancing the performance and stability of the autonomous driving system.

[0053] Example 2

[0054] Figure 2 This is a flowchart of a lane line position detection method provided in Embodiment 2 of the present invention. The relationship between this embodiment and the above embodiments is that the region of interest is determined based on the calculated vehicle turning radius, such as... Figure 2 As shown, the method specifically includes:

[0055] S201. Obtain the vehicle's turning radius based on the vehicle's overall data, and obtain the vehicle's driving curvature based on the vehicle's turning radius, so as to obtain the region of interest (ROI) of the current vehicle based on the vehicle's driving curvature; wherein, if the vehicle's driving curvature is less than or equal to a preset curvature threshold, the left front region and the right front region are taken as ROI; if the vehicle's driving curvature is greater than the preset curvature threshold, and the vehicle's turning direction is left, the left front region and the left rear region are taken as ROI; if the vehicle's driving curvature is greater than the preset curvature threshold, and the vehicle's turning direction is right, the right front region and the right rear region are taken as ROI.

[0056] The wheel turning radius, also known as the turning radius, refers to the radius of the circle traced by the center of the outer front wheel when the vehicle is turning. It measures the vehicle's ability to navigate narrow curves or make U-turns. The turning radius is calculated as the quotient of the front wheel track width and the sine of the front wheel turning angle. Vehicle curvature describes the degree of curvature of the vehicle's trajectory. The vehicle curvature is calculated as the reciprocal of the wheel turning radius. When the vehicle curvature is less than or equal to a preset curvature threshold, it indicates that the vehicle is traveling in a straight line or turning with a small curvature close to a straight line. In this case, the guardrail is usually located at the left or right front of the vehicle, and the left and right front regions are selected as regions of interest.

[0057] When the vehicle's curvature exceeds a preset curvature threshold and the vehicle is turning left, the focus should be on the left front guardrail and left rear guardrail, and the left front and right front areas should be designated as regions of interest. Similarly, when the vehicle's curvature exceeds a preset curvature threshold and the vehicle is turning left, the focus should be on the right front guardrail and right rear guardrail, and the right front and right rear areas should be designated as regions of interest. By using the vehicle's turning radius and curvature, the actual driving state of the vehicle can be determined, enabling targeted acquisition of regions of interest and directional acquisition of detection data in designated areas. This significantly reduces the amount of data processing required for detection and the number of detection radars that need to be activated.

[0058] Optionally, in this embodiment of the invention, the vehicle data includes vehicle speed, steering wheel angle, transmission ratio, front wheel track width, and yaw rate; obtaining the vehicle turning radius based on the vehicle data includes: if it is determined that the vehicle speed is less than a preset speed threshold, obtaining the vehicle turning radius based on the steering wheel angle, the transmission ratio, and the front wheel track width; if it is determined that the vehicle speed is greater than or equal to the preset speed threshold, obtaining the vehicle turning radius based on the vehicle speed and the yaw rate.

[0059] Specifically, at low speeds, i.e., when the vehicle speed is less than a preset speed threshold, the vehicle's steering characteristics closely resemble the Ackermann model. Therefore, the front wheel angle determines the vehicle's steering center. The front wheel angle can be calculated based on the steering wheel angle. For example, when the steering wheel angle is 360 degrees, the corresponding front wheel angle is 30 degrees, meaning the vehicle's gear ratio is 12:1. Based on this, the vehicle's turning radius can be calculated using the following equation:

[0060] ;

[0061] in, Indicates the vehicle's turning radius; This indicates the front wheel track width, which is a fixed parameter in the vehicle, for example... It can be 2.5 meters; This refers to the steering angle of the front wheels.

[0062] At high speeds, i.e., when the vehicle speed is greater than or equal to a preset speed threshold, the vehicle's dynamic characteristics become dominant. The yaw rate reflects the vehicle's rotational rate around its vertical axis and, together with the vehicle speed, determines the instantaneous trajectory curvature. Based on this, the vehicle's turning radius can be calculated using the following equation:

[0063] ;

[0064] in, Indicates the vehicle's turning radius; Indicates the vehicle's speed; This represents the yaw rate; specifically, the sensor detecting the yaw rate can be low-pass filtered to remove high-frequency noise, ensuring the accuracy of the vehicle's turning radius calculation. Therefore, by calculating the vehicle's turning radius at both low and high speeds, the accuracy of the vehicle's trajectory is improved, further enhancing the positioning accuracy of the region of interest.

[0065] S202. Obtain detection data of the region of interest using a detection radar.

[0066] S203. Obtain a candidate guardrail point set based on the detection data, and obtain an associated guardrail point set based on the guardrail point features of the candidate guardrail point set.

[0067] S204. Obtain the guardrail position based on the associated guardrail point set using a multinomial fitting method, and obtain the lane line position corresponding to the current vehicle based on the guardrail position.

[0068] The technical solution of this invention obtains the vehicle's turning radius based on the vehicle's overall data, and obtains the vehicle's driving curvature based on the vehicle's turning radius. Then, it obtains the region of interest (ROI) of the current vehicle based on the vehicle's driving curvature. This achieves targeted acquisition of the ROI and directional acquisition of detection data for a specified area, greatly reducing the amount of data processing required for the detection data. At the same time, by calculating and obtaining the vehicle's turning radius at low and high speeds respectively, the accuracy of the vehicle's driving trajectory is improved, further enhancing the positioning accuracy of the ROI.

[0069] Example 3

[0070] Figure 3 This is a flowchart of a lane line position detection method provided in Embodiment 3 of the present invention. The relationship between this embodiment and the above embodiments is that, based on the Mahalanobis distance between different candidate guardrail points, the set of associated guardrail points is filtered and obtained, such as... Figure 3 As shown, the method specifically includes:

[0071] S301. Obtain the region of interest of the current vehicle based on the vehicle data, and obtain the detection data of the region of interest through the detection radar.

[0072] S302. Obtain a set of candidate guardrail points based on the detection data.

[0073] S303. Based on the position, distance change rate, and angle of each candidate guardrail point, obtain the Mahalanobis distance between different candidate guardrail points.

[0074] Based on the location, distance change rate, and angle of each candidate guardrail point, a four-dimensional covariance matrix is ​​constructed. Substitute the positions of any two candidate guardrail points into the following Mahalanobis distance equation:

[0075] ;

[0076] in, This represents the Mahalanobis distance between the two current candidate guardrail points; This represents the positional difference vector between the two current candidate guardrail points; Let represent the covariance matrix.

[0077] S304. Obtain the set of associated guardrail points based on the Mahalanobis distance between each candidate guardrail point.

[0078] Obtain the Mahalanobis distance between each candidate guardrail point. Then, this value is compared with a pre-set Mahalanobis distance threshold. (For example, 3.0, which corresponds to a 99.7% confidence interval) is compared; if the Mahalanobis distance between the two current candidate guardrail points is less than the Mahalanobis distance threshold, they are judged to be continuous points of the same guardrail and are all added to the associated guardrail point set; based on this, the associated guardrail point set is obtained by calculating the Mahalanobis distance between each candidate guardrail point, which greatly improves the accuracy of obtaining the associated guardrail point set and ensures the accuracy of the guardrail position calculation results.

[0079] S305. Obtain the guardrail position based on the associated guardrail point set using a multinomial fitting method, and obtain the lane line position corresponding to the current vehicle based on the guardrail position.

[0080] Optionally, in this embodiment of the invention, after obtaining the guardrail position based on the associated guardrail point set using a multinomial fitting method, and then obtaining the lane line position corresponding to the current vehicle based on the guardrail position, the method further includes: if a new candidate guardrail point is detected in the candidate guardrail point set, determining whether to add the new candidate guardrail point to the associated guardrail point set based on the Mahalanobis distance between the new candidate guardrail point and each associated guardrail point; if it is determined that the new candidate guardrail point has been added to the associated guardrail point set, updating the guardrail position based on the associated guardrail point set using an incremental fitting method, and then obtaining the lane line position corresponding to the current vehicle based on the updated guardrail position.

[0081] Specifically, the incremental fitting method does not use all associated guardrail points for fitting at once, but rather continuously fits the guardrail points by gradually adding new associated guardrail points. For example, a new associated guardrail point may be added every time the vehicle travels a certain distance, thereby continuously updating the guardrail position. After a new candidate guardrail point appears in the candidate guardrail point set, the covariance matrix of the Mahalanobis distance equation is first updated based on the new candidate guardrail point, and then the Mahalanobis distance between the new candidate guardrail point and each associated guardrail point in the current associated guardrail point set is calculated based on the covariance matrix.

[0082] If the Mahalanobis distance between a newly added candidate guardrail point and all associated guardrail points is less than or equal to the Mahalanobis distance threshold, the new candidate guardrail point is added to the associated guardrail point set. This indicates that the distance between the vehicle and that point is normal and conforms to the distance change trend. Based on this, a cubic equation is refitted according to the associated guardrail point set to update the guardrail position trajectory. If the Mahalanobis distance between a newly added candidate guardrail point and at least one associated guardrail point is greater than the Mahalanobis distance threshold, the new candidate guardrail point is not added to the associated guardrail point set. This indicates that the distance between the vehicle and that point has suddenly increased and does not conform to the distance change trend. It may be a false detection point due to interference with the detection radar. This point is deleted to maintain the consistency of the original associated guardrail point set. Thus, real-time updates of guardrail positions are achieved through incremental fitting. Compared with the traditional single-step fitting method, this avoids interference from abnormal detection points and greatly improves the numerical accuracy of the fitting results.

[0083] The technical solution of this invention obtains the Mahalanobis distance between different candidate guardrail points based on their position, distance change rate, and angle. It then obtains a set of associated guardrail points based on these Mahalanobis distances, significantly improving the accuracy of the associated guardrail point set acquisition and ensuring the accuracy of the guardrail position calculation results. Furthermore, when a new candidate guardrail point is detected in the candidate guardrail point set, the guardrail position is updated based on the Mahalanobis distance between the new candidate guardrail point and each associated guardrail point. Once it is determined that the new candidate guardrail point has been added to the associated guardrail point set, the guardrail position is updated using an incremental fitting method based on the associated guardrail point set. This updated guardrail position allows for the acquisition of the lane line position corresponding to the current vehicle. Thus, the incremental fitting method achieves real-time updating of the guardrail position, avoiding interference from abnormal detection points and greatly improving the numerical accuracy of the fitting results.

[0084] Example 4

[0085] Figure 4 This is a structural block diagram of a lane line position detection device provided in Embodiment 4 of the present invention. The device specifically includes:

[0086] The detection data acquisition module 401 is used to acquire the region of interest of the current vehicle based on the whole vehicle data, and to acquire the detection data of the region of interest through the detection radar;

[0087] The guardrail point set acquisition module 402 is used to acquire a candidate guardrail point set based on the detection data, and acquire an associated guardrail point set based on the guardrail point features of the candidate guardrail point set.

[0088] The guardrail position acquisition module 403 is used to acquire the guardrail position based on the associated guardrail point set using a multinomial fitting method, so as to acquire the lane line position corresponding to the current vehicle based on the guardrail position.

[0089] The technical solution of this invention first obtains the region of interest (ROI) of the current vehicle based on vehicle data, and then acquires detection data of the ROI using a detection radar. Next, it obtains a candidate guardrail point set based on the detection data, and then filters and obtains a set of associated guardrail points. Finally, it obtains the guardrail position based on the guardrail point set using a multinomial fitting method, and then uses the guardrail position to obtain the lane line position corresponding to the current vehicle. This not only achieves real-time detection of lane line positions, but also enables dynamic acquisition of lane line positions based on the real-time guardrail positions. Compared to pre-configured scalar values, this significantly improves the accuracy of lane line detection results, ensuring vehicle driving safety, especially in autonomous driving scenarios, and thus enhancing the performance and stability of the autonomous driving system.

[0090] Optionally, the detection data acquisition module 401 is specifically used to acquire the vehicle's turning radius based on the whole vehicle data, and to acquire the vehicle's driving curvature based on the vehicle's turning radius, so as to acquire the region of interest of the current vehicle based on the vehicle's driving curvature; wherein, if the vehicle's driving curvature is less than or equal to a preset curvature threshold, the left front region and the right front region are taken as regions of interest; if the vehicle's driving curvature is greater than the preset curvature threshold, and the vehicle's turning direction is left, the left front region and the left rear region are taken as regions of interest; if the vehicle's driving curvature is greater than the preset curvature threshold, and the vehicle's turning direction is right, the right front region and the right rear region are taken as regions of interest.

[0091] Optionally, the vehicle data includes vehicle speed, steering wheel angle, gear ratio, front wheel track width, and yaw rate. The detection data acquisition module 401 is further configured to, if it is determined that the vehicle speed is less than a preset speed threshold, obtain the vehicle turning radius based on the steering wheel angle, the gear ratio, and the front wheel track width; and if it is determined that the vehicle speed is greater than or equal to the preset speed threshold, obtain the vehicle turning radius based on the vehicle speed and the yaw rate.

[0092] Optionally, the guardrail point set acquisition module 402 is specifically used to acquire a candidate guardrail point set based on the detection data using a radar cross section threshold and a distance change rate threshold.

[0093] Optionally, the guardrail point set acquisition module 402 is further used to acquire the Mahalanobis distance between different candidate guardrail points based on the position, distance change rate and angle of each candidate guardrail point; and to acquire the associated guardrail point set based on the Mahalanobis distance between each candidate guardrail point.

[0094] Optionally, the lane line position detection device is further configured to, if a new candidate guardrail point is detected in the candidate guardrail point set, determine whether to add the new candidate guardrail point to the associated guardrail point set based on the Mahalanobis distance between the new candidate guardrail point and each associated guardrail point; if it is determined that the new candidate guardrail point has been added to the associated guardrail point set, update the guardrail position based on the associated guardrail point set using an incremental fitting method, so as to obtain the lane line position corresponding to the current vehicle based on the updated guardrail position.

[0095] The above-described apparatus can execute the lane line position detection method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the lane line position detection method provided in any embodiment of the present invention.

[0096] Example 5

[0097] Figure 5 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, electronic devices, blade electronic devices, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0098] like Figure 5As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0099] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0100] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as lane line position detection methods.

[0101] In some embodiments, the lane line position detection method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on a heterogeneous hardware accelerator via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by a processor, one or more steps of the lane line position detection method described above may be performed. Alternatively, in other embodiments, the processor may be configured to perform the lane line position detection method by any other suitable means (e.g., by means of firmware).

[0102] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0103] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0104] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0105] To provide user interaction, the systems and techniques described herein can be implemented on a heterogeneous hardware accelerator, which includes: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the heterogeneous hardware accelerator. Other types of devices can also be used to provide user interaction; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or haptic feedback); and input from the user can be received in any form (including sound input, voice input, or haptic input).

[0106] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0107] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0108] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0109] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A lane line position detection method characterized by, The method comprises the following steps: obtaining a region of interest of the current vehicle according to the whole vehicle data, and obtaining detection data of the region of interest through a detection radar; obtaining a candidate guardrail point set according to the detection data, and obtaining an associated guardrail point set based on a guardrail point feature according to the candidate guardrail point set; obtaining a guardrail position based on a polynomial fitting mode according to the associated guardrail point set, so as to obtain a lane line position corresponding to the current vehicle according to the guardrail position.

2. The lane line position detection method according to claim 1, characterized by, The step of obtaining the region of interest of the current vehicle according to the whole vehicle data comprises the following steps: obtaining a vehicle turning radius according to the whole vehicle data, and obtaining a vehicle running curvature according to the vehicle turning radius, so as to obtain the region of interest of the current vehicle according to the vehicle running curvature; if the vehicle running curvature is less than or equal to a preset curvature threshold, the left front region and the right front region are taken as the region of interest; if the vehicle running curvature is greater than the preset curvature threshold and the vehicle turning direction is left turning, the left front region and the left rear region are taken as the region of interest; if the vehicle running curvature is greater than the preset curvature threshold and the vehicle turning direction is right turning, the right front region and the right rear region are taken as the region of interest.

3. The lane line position detection method according to claim 2, characterized by, The whole vehicle data comprises a vehicle running speed, a steering wheel turning angle, a transmission ratio, a front wheel track and a yaw angular velocity. The step of obtaining the vehicle turning radius according to the whole vehicle data comprises the following steps: if it is determined that the vehicle running speed is less than a preset speed threshold, the vehicle turning radius is obtained according to the steering wheel turning angle, the transmission ratio and the front wheel track; if it is determined that the vehicle running speed is greater than or equal to the preset speed threshold, the vehicle turning radius is obtained according to the vehicle running speed and the yaw angular velocity.

4. The lane line position detection method according to claim 1, characterized by, The step of obtaining the candidate guardrail point set according to the detection data comprises the following step: the candidate guardrail point set is obtained based on a radar scattering cross-section threshold and a distance change rate threshold according to the detection data.

5. The lane line position detection method according to claim 1 or 4, characterized by, The step of obtaining the associated guardrail point set based on the guardrail point feature according to the candidate guardrail point set comprises the following steps: Mahalanobis distances between different candidate guardrail points are obtained according to the positions, distance change rates and angles of the candidate guardrail points; the associated guardrail point set is obtained according to the Mahalanobis distances between the candidate guardrail points.

6. The lane line position detection method according to claim 5, characterized by, After the step of obtaining the guardrail position based on the polynomial fitting mode according to the associated guardrail point set, so as to obtain the lane line position corresponding to the current vehicle according to the guardrail position, the method further comprises the following steps: if it is detected that there is a new candidate guardrail point in the candidate guardrail point set, it is judged whether the new candidate guardrail point is added to the associated guardrail point set according to the Mahalanobis distances between the new candidate guardrail point and each associated guardrail point; if it is determined that the new candidate guardrail point has been added to the associated guardrail point set, the guardrail position is updated based on an incremental fitting mode according to the associated guardrail point set, so as to obtain the lane line position corresponding to the current vehicle according to the updated guardrail position.

7. A lane line position detection device characterized by comprising: The method comprises the following steps: a detection data obtaining module is configured to obtain a region of interest of the current vehicle according to the whole vehicle data, and obtain detection data of the region of interest through a detection radar; a guardrail point set obtaining module is configured to obtain a candidate guardrail point set according to the detection data, and obtain an associated guardrail point set based on a guardrail point feature according to the candidate guardrail point set; The guardrail position acquisition module is configured to acquire a guardrail position based on a polynomial fitting manner according to the associated guardrail point set, so as to acquire a lane line position corresponding to the current vehicle according to the guardrail position.

8. An electronic device, comprising: The electronic device comprises: at least one processor; and a memory connected with the at least one processor in communication; wherein The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the lane line position detection method in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for enabling the processor to implement the lane line position detection method in any one of claims 1-6 when executed. 10.A computer program product comprising a computer program which, when executed by a processor, implements the lane line position detection method in any one of claims 1-6.

Images