Method and system for filtering stop lines based on trajectory

Abstract

The present invention discloses a method and system for trajectory-based stop line filtering. The method obtains a pending stop line and a trajectory within a threshold range around the pending stop line; calculates the angular change of each trajectory point in the trajectory, obtains the trajectory's extreme points based on the angular change, clusters the extreme points using DBSCAN, and extracts trajectory features; finds the base point trajectory point closest to the pending stop line in all trajectories, and determines the correct stop line from the pending stop line based on the trajectory features of all base point trajectory points. The method of trajectory-based stop line filtering according to the present invention can obtain a more reliable stop line, providing better support for autonomous driving behavior decision-making. The method is also simple, fast, and highly robust.

Description

Technical Field

[0001] The present invention relates to a high-precision map production technology, and in particular to a method and system for filtering stop lines based on tracks. Background Art

[0002] In the field of autonomous driving, accurate vehicle control often requires the creation of high-precision maps. Stop line data from these maps can be used to inform autonomous vehicle driving decisions and serve as the end points of lane lines. If an autonomous vehicle's sensors malfunction due to inclement weather such as fog, hail, or heavy rain, the vehicle can rely on the high-precision map data stored within the vehicle for continued driving. High-precision maps also enable autonomous vehicles to process road conditions beyond their normal field of view. However, due to factors such as weather, lighting, and equipment, some stop lines may be misidentified. If these misidentified stop lines are not filtered out, multiple stop lines may appear in the same lane along a stretch of road, significantly reducing driving safety and comfort. Therefore, filtering out misidentified stop lines is necessary to improve safety and comfort.

[0003] To this end, a method for effectively filtering stop lines is needed. Summary of the Invention

[0004] The purpose of the present invention is to overcome the above-mentioned technical deficiencies and propose a method and system for trajectory-based stop line filtering to solve the problem that the existing stop line collection may be misidentified due to factors such as weather, light and equipment, which seriously affects the driving decision-making of autonomous driving vehicles and reduces the safety and comfort of vehicles during driving.

[0005] To achieve the above technical objectives, the technical solution of the present invention provides a method for filtering stop lines based on a trajectory, which comprises the following steps:

[0006] Obtain the pending stop line and the trajectory within the threshold range around the pending stop line;

[0007] Calculate the angle change of each trajectory point in the trajectory, obtain the extreme points of the trajectory based on the angle change of the trajectory point, cluster the extreme points through DBSCAN, and extract the trajectory features;

[0008] Find the base point trajectory point closest to the pending stop line in all trajectories, and determine the correct stop line from the pending stop line based on the trajectory characteristics of all base point trajectory points.

[0009] A second aspect of the present invention provides a system for filtering stop lines based on trajectory, which includes the following functional modules:

[0010] An initial data acquisition module is used to obtain the pending stop line and the trajectory within a threshold range around the pending stop line;

[0011] The feature acquisition module is used to calculate the angle change of each trajectory point in the trajectory, obtain the extreme points of the trajectory based on the angle change of the trajectory point, cluster the extreme points through DBSCAN, and extract the trajectory features;

[0012] The stop line determination module is used to find the base point trajectory point closest to the pending stop line in all trajectories, and determine the correct stop line from the pending stop line based on the trajectory characteristics of all base point trajectory points.

[0013] A third aspect of the present invention provides a server, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for filtering stop lines based on trajectories when executing the computer program.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for filtering stop lines based on trajectories is implemented.

[0015] Compared to existing technologies, the trajectory-based stop line filtering method and system described in the present invention obtains a trajectory within a threshold range around a pending stop line and calculates the angular change of each trajectory point in the trajectory. Based on the angular change, the extreme points of the trajectory are determined. These extreme points are clustered using DBSCAN to extract trajectory features. Finally, the system finds the base trajectory point closest to the pending stop line among all trajectories and determines the correct stop line from the pending stop line based on the trajectory features of all base trajectory points. This trajectory-based stop line filtering method can obtain more reliable stop lines, providing better support for autonomous driving behavior decisions. Furthermore, the method is simple, fast, and highly robust. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flowchart of a method for filtering stop lines based on a trajectory according to an embodiment of the present invention;

[0017] Figure 2 1 is a schematic diagram of segmented trajectory features according to an embodiment of the present invention;

[0018] Figure 3 This is a module block diagram of a system for trajectory filtering stop lines according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0020] like Figure 1 As shown, an embodiment of the present invention provides a method for filtering stop lines based on a trajectory, which includes the following steps:

[0021] S1. Obtain a pending stop line and a trajectory within a threshold range around the pending stop line.

[0022] The trajectory within the threshold range around the pending stop line includes: a trajectory with the center point of the pending stop line as the center and the threshold range as the radius.

[0023] Set the threshold distance between the trajectory point and the pending stop line, set the trajectory obtained from the threshold range around the pending stop line as the initial trajectory, and select the trajectory points whose distance from the pending stop line is less than the threshold distance from each initial trajectory to form a new corrected trajectory.

[0024] S2. Calculate the angle change of each trajectory point in the trajectory, obtain the extreme points of the trajectory based on the angle change of the trajectory point, cluster the extreme points through DBSCAN, and extract the trajectory features.

[0025] The calculation of the angle change of each track point in the track specifically includes:

[0026] Calculate the direction vector of each trajectory point in the modified trajectory;

[0027] Calculate the distance of each track point relative to the previous point in the modified track according to the direction vector of the track point;

[0028] Calculate the deflection angle of each point in the modified trajectory relative to the previous point based on the direction vector of the trajectory point;

[0029] The angle change of the trajectory point whose deflection angle falls within the threshold angle is calculated, and the extreme point of the trajectory is obtained according to the angle change of the trajectory point.

[0030] Clustering extreme points through DBSCAN to extract trajectory features specifically includes:

[0031] Cluster the extreme points using DBSCAN, and select the point with the median index in each class as the representative extreme point of that class, that is, the start and end positions of the corrected trajectory turn, and find their corresponding start and end indexes;

[0032] The trajectory between the start index and the end index is segmented, and the trajectory features of the trajectory points in the segmented trajectory are distinguished according to the trajectory features of the stop line.

[0033] S3. Find the base point track point closest to the undetermined stop line in all the tracks, and determine the correct stop line from the undetermined stop line based on the track features of all the base point track points.

[0034] That is, find the base point trajectory point closest to the undetermined stop line in all trajectories, and determine the correct stop line from the undetermined stop line based on the trajectory characteristics of all base point trajectory points.

[0035] Find the trajectory point closest to the center point of the pending stop line from the initial trajectory, and determine whether the pending stop line is the correct stop line based on the trajectory characteristics of the closest trajectory point. That is, if the trajectory characteristics of the closest trajectory point to the pending stop line in any initial trajectory corresponding to the pending stop line are exiting a turn, then the pending stop line is a misidentified stop line; if the trajectory characteristics of the closest trajectory points to the pending stop line in all initial trajectories corresponding to the pending stop line are not exiting a turn, then the pending stop line is the correct stop line.

[0036] The specific implementation steps of the method based on trajectory filtering stop line are as follows:

[0037] 1) Data Acquisition

[0038] 1.1. The collected pending stop line data is a line segment represented by two points, with the two endpoints being A and B. Assume that point A is the left endpoint and point B is the right endpoint;

[0039] 1.2. The collected initial trajectory data is a set of multiple initial trajectories {T1, T2, T3, ..., T n}, n is the number of trajectories, each initial trajectory T i A set of trajectory points arranged in time order m i is the number of trajectory points of the i-th trajectory.

[0040] 2) Trajectory feature extraction

[0041] 2.1. The center point C of the undetermined stop line is the midpoint between endpoints A and B, C = (A + B) / 2;

[0042] 2.2. Calculate the midpoint C and trajectory T i The distance d of each trajectory point in the jth trajectory point is d j =||CP j ||, get the distance set D j ;

[0043] 2.3. From set D jSelect the distance d j For trajectory points ≤100m, if all d j If it is greater than 100, a new initial trajectory T is selected. k , execute step 2.2; if there is d j For trajectory points no greater than 100m, the trajectory points that meet the conditions are extracted in sequence to form a new corrected trajectory T′ i , the number of points is m′ i .

[0044] 2.4. Calculate the corrected trajectory T′ i The direction vector of each trajectory point in is, when j=1, v1=(0,0), when j>1, the vector v of the jth trajectory point is j =P j -P j-1 ,

[0045] 2.5. Calculate the corrected trajectory T′ i The distance between each trajectory point and the previous trajectory point is, when j = 1, d′1 = 0, when j> 1, the vector d′ of the jth trajectory point is j =|v j |,

[0046] 2.6. Calculate the corrected trajectory T′ i The deflection angle α of each trajectory point relative to the previous trajectory point, when j≤2, α j = 0, when j>2, the deflection angle vector α of the jth trajectory point j =cos -1 ((v j ·v j-1 ) / (||v j ||*||v j-1 ||)) Calculate the angular deviation of each trajectory point relative to the starting point If there is e j ≥80°, execute step 2.7; otherwise, reselect an initial trajectory T k , proceed to step 2.2;

[0047] 2.7. Calculate the angle change ρ of each trajectory point. When j = 1, ρ1 = ρ2. When j > 1, ρ j =α j / d j The turning point is the extreme point of ρ. If ρ j *ρ j+1 ≤0 and e j ≥80, then point P j That is the extreme point, if ρ j *ρ j+1 ≤0 and the 10 trajectory points around the jth point correspond to ej There is e j ≥30, then point P j This is the extreme point. Get a series of extreme points

[0048] 2.8. Cluster the extreme points using DBSCAN, with a minimum number of 1 per class and a cluster radius of 1m. The clustering results are as follows: Figure 2 As shown in the circles, the points with the median index in each category are the representative extreme points of this category, which are and That is, the starting position and the ending position of the trajectory turning, the corresponding indexes are s and e, such as Figure 1 As shown by the red dots. Divide s to e into 4 segments according to the index, and the index interval of each segment is x is rounded downwards to an integer, and the segmentation point indices are se1 = s + x, se2 = s + 2 * x, and se3 = s + 3 * x.

[0049] 2.9. Correct the trajectory T′ i The trajectory feature of the trajectory points between the starting trajectory point and the (s-1) point of the corrected trajectory is defined as 1 (straight forward), the trajectory feature of the trajectory points between s and se1 is defined as 2 (entering a turn), the trajectory feature of the trajectory points between se1 and se3 is defined as 3 (turning), the trajectory feature of the trajectory points between se3 and e is defined as 4 (exiting a turn), and the trajectory feature of the trajectory points from se3 to the last trajectory point is defined as 1 (straight forward);

[0050] 3) Stop line filtering

[0051] 3.1. Calculate the center point C of the undetermined stop line and the initial trajectory T i The distance d between each trajectory point j =||CP j ||, get the distance set D j , find the closest trajectory point

[0052] 3.2、If any of the initial trajectories corresponding to the pending stop line, The trajectory feature is 4 (out of the turn), then the pending stop line is the misidentified stop line; if among all the initial trajectories corresponding to the pending stop line, If the trajectory characteristics of are not 4 (exiting the turn), then the undetermined stop line is the correct stop line.

[0053] The present invention obtains a trajectory within a threshold range around a pending stop line and calculates the angular change of each trajectory point. Based on this angular change, the extreme points of the trajectory are determined. These extreme points are clustered using DBSCAN to extract trajectory features. Finally, the base point closest to the pending stop line is found among all trajectories. Based on the trajectory features of all base points, the correct stop line is determined from the pending stop line. The proposed trajectory-based stop line filtering method can produce more reliable stop lines, providing better support for autonomous driving behavior decisions. The method is simple, fast, and highly robust.

[0054] like Figure 3 As shown, an embodiment of the present invention further discloses a system for filtering stop lines based on trajectory, which includes the following functional modules:

[0055] An initial data acquisition module 10 is used to acquire a pending stop line and a trajectory within a threshold range around the pending stop line;

[0056] The feature acquisition module 20 is used to calculate the angle change of each trajectory point in the trajectory, obtain the extreme points of the trajectory based on the angle change of the trajectory points, cluster the extreme points through DBSCAN, and extract the trajectory features;

[0057] The stop line determination module 30 is used to find the base point track point closest to the pending stop line in all trajectories, and determine the correct stop line from the pending stop line based on the trajectory characteristics of all the base point track points.

[0058] The execution method of the system for filtering stop lines based on trajectory in this embodiment is substantially the same as the above-mentioned method for filtering stop lines based on trajectory, and thus will not be described in detail.

[0059] The server in this embodiment is a device that provides computing services, typically a computer with high computing power that is provided to multiple consumers via a network. The server in this embodiment includes memory, a processor, and a system bus. The memory includes executable programs stored thereon. Those skilled in the art will appreciate that the terminal device structure in this embodiment does not limit the terminal device and may include more or fewer components than shown, or combinations of certain components, or different component arrangements.

[0060] The memory can be used to store software programs and modules. The processor executes the various functional applications and data processing of the terminal by running the software programs and modules stored in the memory. The memory can mainly include a program storage area and a data storage area. The program storage area can store an operating system and at least one application required for a function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created based on the use of the terminal (such as audio data, a phone book, etc.). In addition, the memory can include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0061] An executable program for a method for trajectory-based stop line filtering is included in a memory. The executable program can be divided into one or more modules / units, which are stored in the memory and executed by a processor to complete the information acquisition and implementation process. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which are used to describe the execution process of the computer program in the server. For example, the computer program can be divided into an initial data acquisition module 10, a feature acquisition module 20, and a stop line determination module 30.

[0062] The processor is the control center of the server, connecting the various components of the entire terminal device using various interfaces and lines. By running or executing software programs and / or modules stored in memory and accessing data stored in memory, it performs various terminal functions and processes data, thereby providing overall terminal monitoring. Optionally, the processor may include one or more processing units; preferably, the processor may integrate an application processor and a modem processor, wherein the application processor primarily processes the operating system, application programs, etc., while the modem processor primarily handles wireless communications. It is understood that the modem processor may not be integrated into the processor.

[0063] The system bus connects the various functional components within a computer and transmits data, address information, and control information. It can be a PCI bus, ISA bus, or VESA bus. Processor instructions are transmitted to memory via the bus, and memory feeds data back to the processor. The system bus is responsible for the exchange of data and instructions between the processor and memory. Of course, the system bus can also connect to other devices, such as network interfaces and display devices.

[0064] The server should at least include a CPU, a chipset, a memory, a disk system, etc. Other components will not be described in detail here.

[0065] In an embodiment of the present invention, the executable program executed by the processor included in the terminal is specifically: a method for filtering stop lines based on a trajectory, which includes the following steps:

[0066] Obtain the pending stop line and the trajectory within the threshold range around the pending stop line;

[0067] Calculate the angle change of each trajectory point in the trajectory, obtain the extreme points of the trajectory based on the angle change of the trajectory point, cluster the extreme points through DBSCAN, and extract the trajectory features;

[0068] Find the base point trajectory point closest to the pending stop line in all trajectories, and determine the correct stop line from the pending stop line based on the trajectory characteristics of all base point trajectory points.

[0069] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0070] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0071] Those skilled in the art will appreciate that the modules, units, and / or method steps of the various embodiments described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0072] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for filtering stop lines based on trajectory, characterized in that: The steps include: Obtain the pending stop line and the trajectory within the threshold range around the pending stop line; Calculate the angle change of each trajectory point in the trajectory, obtain the extreme points of the trajectory based on the angle change of the trajectory point, cluster the extreme points through DBSCAN, and extract the trajectory features; Find the base point closest to the pending stop line among all the trajectories, and determine the correct stop line from the pending stop line based on the trajectory characteristics of all the base points; The process of finding the closest base point to the pending stop line among all the trajectories and determining the correct stop line from the pending stop line based on the trajectory characteristics of all the base points includes the following: According to the trajectory characteristics of all base point trajectory points, the correct stop line is determined from the undetermined stop line; Find the trajectory point closest to the center point of the pending stop line in the initial trajectory, and determine whether the pending stop line is the correct stop line based on the trajectory characteristics of the closest trajectory point. That is, if the trajectory characteristics of the closest trajectory point to the pending stop line in any initial trajectory corresponding to the pending stop line are out of turn, then the pending stop line is a misidentified stop line; If the trajectory features of all the initial trajectories corresponding to the pending stop line and the nearest trajectory points to the pending stop line are not exit turns, then the pending stop line is the correct stop line.

2. The method for filtering stop lines based on trajectory according to claim 1, characterized in that: The trajectory within the threshold range around the pending stop line includes: a trajectory with the center point of the pending stop line as the center and the threshold range as the radius.

3. The method for filtering stop lines based on trajectory according to claim 1, characterized in that: Before calculating the angle change of each track point in the trajectory, the trajectory needs to be corrected. That is, a threshold distance is set between the track point and the pending stop line. The track obtained from the threshold range around the pending stop line is set as the initial track. From each initial track, track points whose distance from the pending stop line is less than the threshold distance are selected to form a new corrected track.

4. The method for filtering stop lines based on trajectory according to claim 3, characterized in that: The calculation of the angle change of each track point in the track includes the following: Calculate the direction vector of each trajectory point in the corrected trajectory; Calculate the distance of each trajectory point relative to the previous point in the corrected trajectory according to the direction vector of the trajectory point; Calculate the deflection angle of each point in the corrected trajectory relative to the previous point based on the direction vector of the trajectory point; The angle change of the trajectory point whose deflection angle falls within the threshold angle is calculated, and the extreme point of the trajectory is obtained according to the angle change of the trajectory point.

5. The method for filtering stop lines based on trajectory according to claim 4, characterized in that: The extreme points are clustered by DBSCAN to extract trajectory features, including the following: Cluster the extreme points using DBSCAN, and select the point with the median index in each class as the representative extreme point of that class, that is, the start and end positions of the corrected trajectory turn, and find their corresponding start and end indexes; The trajectory between the start index and the end index is segmented, and the trajectory features of the trajectory points in the segmented trajectory are distinguished according to the trajectory features of the stop line.

6. The method for filtering stop lines based on trajectory according to claim 5, characterized in that: The segmenting of the trajectory between the start index and the end index, and distinguishing the trajectory features of the trajectory points in the segmented trajectory according to the trajectory features of the stop line, specifically includes: Divide the trajectory between the start index s and the end index e into 4 segments, and get the indexes of the 3 segment points respectively 、 、 ; The trajectory features of the trajectory points between the starting point of the corrected trajectory and the point (s-1) are defined as straight lines, from s to The trajectory features of the trajectory points between points are defined as entering a turn, from arrive The trajectory features of the trajectory points between points are defined as turns, from The trajectory features of the trajectory points between point e and point e are defined as turning out, and the trajectory features of the trajectory points between point e and the last trajectory point are defined as straight ahead.

7. A system based on trajectory filtering stop line, characterized in that: Includes the following functional modules: An initial data acquisition module is used to obtain the pending stop line and the trajectory within a threshold range around the pending stop line; The feature acquisition module is used to calculate the angle change of each trajectory point in the trajectory, obtain the extreme points of the trajectory based on the angle change of the trajectory point, cluster the extreme points through DBSCAN, and extract the trajectory features; The stop line determination module is used to find the base point trajectory point closest to the undetermined stop line in all trajectories, and determine the correct stop line from the undetermined stop line based on the trajectory characteristics of all the base point trajectory points; The process of finding the closest base point to the pending stop line among all the trajectories and determining the correct stop line from the pending stop line based on the trajectory characteristics of all the base points includes the following: According to the trajectory characteristics of all base point trajectory points, the correct stop line is determined from the undetermined stop line; Find the trajectory point closest to the center point of the pending stop line in the initial trajectory, and determine whether the pending stop line is the correct stop line based on the trajectory characteristics of the closest trajectory point. That is, if the trajectory characteristics of the closest trajectory point to the pending stop line in any initial trajectory corresponding to the pending stop line are out of turn, then the pending stop line is a misidentified stop line; If the trajectory features of all the initial trajectories corresponding to the pending stop line and the nearest trajectory points to the pending stop line are not exit turns, then the pending stop line is the correct stop line.

8. A server comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for filtering stop lines based on trajectories according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for filtering stop lines based on trajectories according to any one of claims 1 to 6 is implemented.

Images