CN118230276A

Movatterモバイル変換

Info

Publication number: CN118230276A
Application number: CN202410663421.6A
Authority: CN
Inventors: 石创开; 宁尚尚; 朱清广; 孙川; 郑伟克; 马圣策; 姜海鹏
Original assignee: Great Wall Motor Co Ltd
Current assignee: Great Wall Motor Co Ltd
Priority date: 2024-05-27
Filing date: 2024-05-27
Publication date: 2024-06-21
Anticipated expiration: 2044-05-27
Also published as: CN118230276B

Abstract

The application provides a method, a device, a cloud server and a storage medium for detecting a lane crossing point, wherein the method is applied to the field of vehicle driving and comprises the following steps: acquiring at least one first data point corresponding to a first lane line of a first road section and a plurality of second data points corresponding to a second lane line of the first road section from map data; determining whether at least one candidate intersection exists between the first lane line and the second lane line based on the at least one first data point and the plurality of second data points; in the case where there is at least one candidate intersection between the first lane line and the second lane line, a target intersection between the first lane line and the second lane line is determined based on the at least one candidate intersection, the target intersection being used to represent an actual intersection position of the first lane line and the second lane line. The method can simply and efficiently identify the intersection point of two lane lines, and provides accurate road information for vehicle running.

Description

Method and device for detecting lane crossing point, cloud server and storage medium

Technical Field

The present application relates to the field of vehicle driving, and more particularly, to a method, apparatus, cloud server, and storage medium for detecting a lane crossing in the field of vehicle driving.

Background

Currently, in order to recommend a precise driving route to a user during driving of a vehicle, acquisition and updating of road information become necessary preconditions.

Along with the development of intelligent driving technology, in order to obtain the road information of each road in the running process of the vehicle, a plurality of vehicles running on the road can be adopted to collect the road information and upload the road information to a cloud end (or a cloud server), and the cloud end analyzes and processes the road information to obtain a final high-precision map. Such high-precision maps collectively constructed from information collected by a plurality of vehicles are also referred to as "crowd-sourced maps" or crowd-sourced maps.

In order to ensure safe driving of the vehicle, detection of lane lines becomes a necessary premise in the process of constructing a crowd source map. In the lane line detection process, there may be a case where a plurality of lane lines merge into or merge out from each other, that is, there is an intersection between the plurality of lane lines. Therefore, how to detect the intersection of lane lines becomes a problem to be solved.

Disclosure of Invention

The application provides a method, a device, a cloud server and a storage medium for detecting a lane crossing point.

In the above technical solution, a method for detecting a lane crossing is provided, where at least one first data point corresponding to a first lane line of a first road segment and a plurality of second data points corresponding to a second lane line of the first road segment are obtained from map data obtained in advance, where the first lane line is adjacent to the second lane line, and when the two lane lines are parallel, the driving directions of the two lane lines are the same. Based on the at least one first data point and the plurality of second data points, it is determined whether at least one candidate intersection exists between the first lane line and the second lane line. The candidate intersection can indicate that the probability of intersection of the first lane line and the second lane line is relatively large. Therefore, the process can realize the pre-judgment of whether the lane lines are crossed or not, and accurately identify the relative position relationship between the two lane lines. When at least one candidate intersection exists between the first lane line and the second lane line, a target intersection between the first lane line and the second lane line is determined. Under the condition that two lane lines are crossed, the final target crossing point is obtained through the candidate crossing point, and the lane line crossing point can be accurately identified.

With reference to the first aspect, in some possible implementations, determining whether at least one candidate intersection exists between the first lane line and the second lane line according to the at least one first data point and the plurality of second data points includes: judging whether the change rate of the vertical distance between the first lane line and the second lane line is greater than a preset change rate according to the at least one first data point and the plurality of second data points; when the change rate of the vertical distance is larger than the preset change rate, starting from the end position of the first road section, acquiring a plurality of third data points corresponding to the first road line of a second road section and a plurality of fourth data points corresponding to the second road line of the second road section from the map data, wherein the direction of the second road section is the same as the driving direction; determining whether the at least one candidate intersection exists between the first lane line and the second lane line based on the plurality of third data points and the plurality of fourth data points.

In the above technical scheme, in the crossing scene, for the lane line merging, two adjacent lane lines keep a parallel relationship before merging, and the vertical distance between the two lane lines basically keeps unchanged. For lane line merging, the two lane lines substantially coincide and the vertical distance between the two lane lines remains substantially unchanged prior to merging. When two lane lines intersect, the two lane lines become non-parallel or no longer coincide when approaching the intersection, and the vertical distance changes. Therefore, when determining whether the candidate intersection exists between two adjacent lane lines, it is possible to determine whether a point at which the vertical distance is suddenly changed, that is, a point at which the change rate of the vertical distance is greater than a preset change rate, exists on any one lane line. If so, a plurality of third data points are respectively taken on the first lane line and a plurality of fourth data points are taken on the second lane line along the driving directions of the two lane lines from the current point. The points with abrupt change of the distance indicate the possibility of intersection of the two lane lines, so that the rapid pre-judgment on whether the two lane lines intersect or not is realized through the change rate of the vertical distance. When a crossing is likely to exist, a third data point and a fourth data point are acquired to determine whether a candidate crossing exists. Because the vertical distance between the two lane lines is basically unchanged before mutation, the data points before mutation are selected to judge whether candidate crossing points exist or not, and the meaning is not great and the result is inaccurate. Therefore, the process can ensure the reliability and the accuracy of data point selection, and ensure the accuracy of candidate intersection judgment.

With reference to the first aspect and the foregoing implementation manner, in some possible implementation manners, the determining, according to the third data points and the fourth data points, whether the at least one candidate intersection point exists between the first lane line and the second lane line includes: determining a plurality of first vertical distances of the first lane line and the second lane line from the plurality of third data points and the plurality of fourth data points; determining a distance change trend according to the plurality of first vertical distances, wherein the distance change trend is used for representing the change rule of the vertical distances along the driving direction; determining whether the at least one candidate intersection exists between the first lane line and the second lane line according to the distance variation trend and/or the plurality of first vertical distances.

With reference to the first aspect and the foregoing implementation manner, in some possible implementation manners, the determining whether the at least one candidate intersection point exists between the first lane line and the second lane line according to the distance variation trend and/or the plurality of first vertical distances includes: determining that the at least one candidate intersection exists between the first lane line and the second lane line in the case where the distance variation trend satisfies a preset trend and at least one first vertical distance less than or equal to a first threshold exists among the plurality of first vertical distances; and determining that the at least one candidate intersection point does not exist between the first lane line and the second lane line in the case that the distance variation trend does not satisfy the preset trend or the plurality of first vertical distances are all greater than the first threshold.

Optionally, the preset trend includes increasing or decreasing.

In the above technical solution, when determining whether at least one candidate intersection exists between the first lane line and the second lane line, a plurality of first vertical distances between the two lane lines are first determined based on a plurality of third data points and a plurality of fourth data points. From the plurality of first vertical distances, a distance variation trend, i.e. in particular a gradual increase in distance or a gradual decrease in distance or neither a gradual increase nor a gradual decrease, may be obtained. When the distance change trend is gradually increasing or gradually decreasing, it is indicated that the two lane lines satisfy the intersecting condition. After the intersection condition is satisfied, since the vertical distance between the two lane lines is small near the intersection point, it is also required to determine whether the vertical distance less than or equal to the first threshold value is any of the plurality of first vertical distances, that is, whether the third data point and the fourth data point currently selected are already close to the intersection point after the two lane lines satisfy the intersection condition, and if so, it is described that the two lane lines are about to merge or exit. The process can accurately distinguish whether the lane lines are about to be merged or separated through the distance change trend, and can ensure the accuracy of determining the intersection point of the two lane lines.

With reference to the first aspect and the foregoing implementation manners, in some possible implementation manners, after determining that the at least one candidate intersection point exists between the first lane line and the second lane line, the method further includes: the at least one candidate intersection is determined as at least one third data point of the plurality of third data points corresponding to the at least one first vertical distance and at least one fourth data point of the plurality of fourth data points corresponding to the at least one first vertical distance.

In the above technical solution, at least one first vertical distance is less than or equal to a first threshold value, which indicates that the first lane line and the second lane line are infinitely close. Thus, the present application may consider a third data point of the plurality of third data points for determining the at least one first vertical distance, and a fourth data point of the plurality of fourth data points for determining the at least one first vertical distance as candidate intersection points, i.e., some data points when the first lane line and the second lane line almost intersect. The above procedure can thus ensure the accuracy and reliability of candidate intersection selection.

With reference to the first aspect and the foregoing implementation manner, in some possible implementation manners, the preset trend includes increasing or decreasing, and determining, in a case where the at least one candidate intersection exists between the first lane line and the second lane line, a target intersection between the first lane line and the second lane line according to the at least one candidate intersection includes: fusing the at least one candidate intersection to obtain a predicted intersection; when the distance change trend is reduced, starting with the predicted intersection point, acquiring a plurality of fifth data points corresponding to the first road line of a third road section and a plurality of sixth data points corresponding to the second road line of the third road section, wherein the direction of the third road section is the same as the driving direction; determining the target intersection based on the predicted intersection, the plurality of fifth data points, and the plurality of sixth data points; when the distance change trend is increased, starting with the predicted intersection point, acquiring a plurality of seventh data points corresponding to the first lane line of a fourth road section and a plurality of eighth data points corresponding to the second lane line of the fourth road section, wherein the direction of the fourth road section is opposite to the driving direction; the target intersection is determined based on the predicted intersection, the plurality of seventh data points, and the plurality of eighth data points.

In the above technical solution, after obtaining at least one candidate intersection, in order to obtain a final target intersection, at least one candidate intersection is first fused to obtain a predicted intersection. In consideration of the stability of the intersection of the first lane line and the second lane line, after the predicted intersection is obtained, a plurality of data points of the two lane lines within a certain distance can be obtained to determine whether the two lane lines also exhibit a stable intersection trend. According to the difference of the distance change trend, the method is concretely divided into the following two scenes:

The first is to indicate that two lane lines may merge in the direction of travel as the distance gradually decreases. After the predicted intersection is determined, a target intersection which merges into the lane line is determined according to the predicted intersection and a plurality of fifth data points and a plurality of sixth data points which correspond to the two lane lines in the third road section along the driving direction.

The second is to indicate that the two lane lines may merge in the direction of travel when the distance increases gradually. After the predicted intersection is determined, a target intersection at which the lane lines merge is determined in a fourth link in the opposite direction to the traveling direction based on the predicted intersection and a plurality of seventh data points and a plurality of eighth data points corresponding to the two lane lines in the fourth link, respectively.

The process can flexibly determine the intersection point of two lane lines under different intersection scenes, and can ensure the stability and accuracy of the determination of the target intersection point.

With reference to the first aspect and the foregoing implementation manner, in some possible implementation manners, the determining the target intersection according to the predicted intersection, the plurality of fifth data points, and the plurality of sixth data points includes: determining a plurality of second vertical distances from the plurality of fifth data points and the plurality of sixth data points; determining the target intersection as the predicted intersection if the plurality of second vertical distances are each less than or equal to a second threshold; and determining the target intersection from the predicted intersection, the plurality of seventh data points, and the plurality of eighth data points, comprising: determining a plurality of third vertical distances from the plurality of seventh data points and the plurality of eighth data points; and determining the target intersection as the predicted intersection when the plurality of third vertical distances are all less than or equal to the second threshold.

In the above technical solution, after the predicted intersection is obtained when two lane lines are converged, the present application may further determine a plurality of second vertical distances in a third path segment after the predicted intersection according to a plurality of fifth data points and a plurality of sixth data points, and determine that the predicted intersection is the target intersection when the plurality of second vertical distances are all less than or equal to a second threshold. Similarly, when two lane lines merge, a plurality of third vertical distances in the fourth road section after the predicted intersection point can be determined according to the seventh data points and the eighth data points, and when the third vertical distances are smaller than or equal to the second threshold value, the predicted intersection point is determined to be the target intersection point. The process can ensure the stability of the intersection of the lane lines and avoid inaccurate determination of the intersection caused by the fact that only partial areas of the two lane lines intersect.

With reference to the first aspect and the foregoing implementation manners, in some possible implementation manners, after determining that the at least one candidate intersection point exists between the first lane line and the second lane line, the method further includes: and determining the crossing type of the first lane line and the second lane line according to the distance change trend, wherein the crossing type is used for representing the crossing direction of the first lane line and the second lane line.

With reference to the first aspect and the foregoing implementation manner, in some possible implementation manners, the preset trend includes increasing or decreasing, the intersection type includes a lane line junction or a lane line junction, and determining, according to the distance change trend, the intersection type of the first lane line and the second lane line includes: determining that the intersection type is lane line merging under the condition that the distance change trend is reduced; in the case where the distance change trend is increasing, the intersection type is determined as a lane line junction.

In the above technical solution, the present application further provides a method for determining whether two lane lines are converging or converging based on a distance change trend between the two lane lines, where the two lane lines are converging when the distance change trend along the driving direction is reduced; when the distance change trend increases in the traveling direction, it means that the two lane lines are converging. Therefore, the scheme of the application not only can determine the intersection point of the two lane lines, but also can simply, efficiently and accurately identify the intersection type of the two lane lines.

In a second aspect, there is provided an apparatus for detecting a lane line intersection, the apparatus comprising: the data acquisition module is used for acquiring at least one first data point corresponding to a first lane line of a first road segment and a plurality of second data points corresponding to a second lane line of the first road segment from map data, wherein the second lane line is a lane line adjacent to the first lane line, and when the first lane line is parallel to the second lane line, the running directions of the first lane line and the second lane line are the same; a candidate intersection judging module, configured to determine, according to the at least one first data point and the plurality of second data points, whether at least one candidate intersection exists between the first lane line and the second lane line, where a probability of intersection between the first lane line and the second lane line is greater than a preset probability; a target intersection determination module for determining a target intersection between the first lane line and the second lane line based on the at least one candidate intersection in the presence of the at least one candidate intersection between the first lane line and the second lane line, the target intersection being indicative of an actual intersection position of the first lane line and the second lane line.

With reference to the second aspect, in some possible implementations, the candidate intersection determination module is specifically configured to: judging whether the change rate of the vertical distance between the first lane line and the second lane line is greater than a preset change rate according to the at least one first data point and the plurality of second data points; when the change rate of the vertical distance is larger than the preset change rate, starting from the end position of the first road section, acquiring a plurality of third data points corresponding to the first road line of a second road section and a plurality of fourth data points corresponding to the second road line of the second road section from the map data, wherein the direction of the second road section is the same as the driving direction; determining whether the at least one candidate intersection exists between the first lane line and the second lane line based on the plurality of third data points and the plurality of fourth data points.

With reference to the second aspect and the foregoing implementation manner, in some possible implementation manners, the candidate intersection judgment module is further configured to: determining a plurality of first vertical distances of the first lane line and the second lane line from the plurality of third data points and the plurality of fourth data points; determining a distance change trend according to the plurality of first vertical distances, wherein the distance change trend is used for representing the change rule of the vertical distances along the driving direction; determining whether the at least one candidate intersection exists between the first lane line and the second lane line according to the distance variation trend and/or the plurality of first vertical distances.

With reference to the second aspect and the foregoing implementation manner, in some possible implementation manners, the candidate intersection judgment module is further configured to: determining that the at least one candidate intersection exists between the first lane line and the second lane line in the case where the distance variation trend satisfies a preset trend and at least one first vertical distance less than or equal to a first threshold exists among the plurality of first vertical distances; and determining that the at least one candidate intersection point does not exist between the first lane line and the second lane line in the case that the distance variation trend does not satisfy the preset trend or the plurality of first vertical distances are all greater than the first threshold.

With reference to the second aspect and the foregoing implementation manners, in some possible implementation manners, after determining that the at least one candidate intersection point exists between the first lane line and the second lane line, the apparatus further includes: a candidate intersection determination module for determining the at least one candidate intersection as at least one third data point of the plurality of third data points corresponding to the at least one first vertical distance and at least one fourth data point of the plurality of fourth data points corresponding to the at least one first vertical distance.

With reference to the second aspect and the foregoing implementation manner, in some possible implementation manners, the preset trend includes increasing or decreasing, and the target intersection determining module is specifically configured to: fusing the at least one candidate intersection to obtain a predicted intersection; when the distance change trend is reduced, starting with the predicted intersection point, acquiring a plurality of fifth data points corresponding to the first road line of a third road section and a plurality of sixth data points corresponding to the second road line of the third road section, wherein the direction of the third road section is the same as the driving direction; determining the target intersection based on the predicted intersection, the plurality of fifth data points, and the plurality of sixth data points; when the distance change trend is increased, starting with the predicted intersection point, acquiring a plurality of seventh data points corresponding to the first lane line of a fourth road section and a plurality of eighth data points corresponding to the second lane line of the fourth road section, wherein the direction of the fourth road section is opposite to the driving direction; the target intersection is determined based on the predicted intersection, the plurality of seventh data points, and the plurality of eighth data points.

With reference to the second aspect and the foregoing implementation manner, in some possible implementation manners, the target intersection determining module is further configured to: determining a plurality of second vertical distances from the plurality of fifth data points and the plurality of sixth data points; determining the target intersection as the predicted intersection if the plurality of second vertical distances are each less than or equal to a second threshold; and determining a plurality of third vertical distances from the plurality of seventh data points and the plurality of eighth data points; and determining the target intersection as the predicted intersection when the plurality of third vertical distances are all less than or equal to the second threshold.

With reference to the second aspect and the foregoing implementation manners, in some possible implementation manners, after determining that the at least one candidate intersection point exists between the first lane line and the second lane line, the apparatus further includes: and the intersection type determining module is used for determining the intersection type of the first lane line and the second lane line according to the distance change trend, wherein the intersection type is used for representing the intersection direction of the first lane line and the second lane line.

With reference to the second aspect and the foregoing implementation manner, in some possible implementation manners, the preset trend includes increasing or decreasing, the intersection type includes lane line merging or lane line merging, and the intersection type determining module is specifically configured to: determining that the intersection type is lane line merging under the condition that the distance change trend is reduced; in the case where the distance change trend is increasing, the intersection type is determined as a lane line junction.

In a third aspect, a cloud server is provided that includes a memory and a processor. The memory is for storing executable program code, and the processor is for calling and running the executable program code from the memory, so that the cloud server performs the method of the first aspect or any of the possible implementation manners of the first aspect.

In a fourth aspect, there is provided a computer program product comprising: computer program code which, when run on a computer, causes the computer to perform the method of the first aspect or any one of the possible implementations of the first aspect.

In a fifth aspect, a computer readable storage medium is provided, the computer readable storage medium storing computer program code which, when run on a computer, causes the computer to perform the method of the first aspect or any one of the possible implementations of the first aspect.

Drawings

Fig. 1 is a schematic view of a scene of acquiring map data according to an embodiment of the present application;

FIG. 2 is a schematic flow chart of a method for detecting lane crossing points provided by an embodiment of the present application;

FIG. 3 (a) is a schematic view of a scenario in which data points are acquired according to an embodiment of the present application;

FIG. 3 (b) is a schematic view of another scenario in which data points are acquired according to an embodiment of the present application;

FIG. 4 (a) is a schematic view of a scenario for calculating a vertical distance according to an embodiment of the present application;

FIG. 4 (b) is a schematic view of another scenario for calculating a vertical distance according to an embodiment of the present application;

Fig. 5 is a schematic structural diagram of a device for detecting a lane crossing according to an embodiment of the present application;

fig. 6 is a schematic structural diagram of a cloud server according to an embodiment of the present application.

Detailed Description

The technical scheme of the application will be clearly and thoroughly described below with reference to the accompanying drawings. Wherein, in the description of the embodiments of the present application, unless otherwise indicated, "/" means or, for example, a/B may represent a or B: the text "and/or" is merely an association relation describing the associated object, and indicates that three relations may exist, for example, a and/or B may indicate: the three cases where a exists alone, a and B exist together, and B exists alone, and furthermore, in the description of the embodiments of the present application, "plural" means two or more than two.

The terms "first," "second," and the like, are used below for descriptive purposes only and are not to be construed as implying or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature.

A scene of acquiring map data in the crowdsourcing mode is described below.

Fig. 1 is a schematic view of a scene of acquiring map data according to an embodiment of the present application.

Illustratively, as shown in fig. 1, the vehicle 102 and the vehicle 104 log into the cloud server 105 and are communicatively connected to the cloud server 105 by means of authentication, respectively. After the communication connection is established, when map data is acquired in the crowdsourcing mode, the vehicle 102 may continuously collect data of various objects on the road during the driving process and upload the data to the cloud server 105. Similarly, the vehicle 104 may continuously collect data of various objects on the road during the driving process, and upload the data to the cloud server 105. After receiving the data sent by the vehicle 102 and the vehicle 104, the cloud server 105 may process a large amount of data to obtain the latest high-precision map.

In a possible implementation, when the vehicle 102 travels to the lane corresponding to the lane line 101, data of the lane line 101 may be acquired during traveling and sent to the cloud server 105.

In another possible implementation, when the vehicle 104 travels to the lane corresponding to the lane line 103, the data of the lane line 103 may be acquired during the traveling process and sent to the cloud server 105.

The cloud server 105 may acquire and process data of a plurality of lane lines on the road in the above manner.

For example, as shown in fig. 1, when the lane line 101 and the lane line 103 intersect, in the processing procedure, the cloud server 105 may determine, based on the acquired data of the lane line 101 and the data of the lane line 103, the intersection of the lane line 101 and the lane line 103 in order to ensure accuracy of data issued to each vehicle.

At present, in the related art, an image of a lane line is obtained, feature extraction is performed on the image to obtain a feature image, and finally the feature image is input into a cross point detection model, so that a cross point of the lane line is obtained.

The lane line crossing points are obtained through image processing, and the problem of inaccurate image acquisition is unavoidable. In addition, the image is subjected to feature extraction and then input into the cross point model, so that the period of the determining process of the cross point is long and the process is complicated.

Based on the problems, the application provides a method for detecting the intersection point of the lane lines, which can simply and efficiently identify the intersection point of two lane lines and provides accurate road information for vehicle running.

After the application scenario of the embodiment of the present application is introduced, a method for detecting a lane crossing point provided by the embodiment of the present application is described below.

Fig. 2 is a schematic flow chart of a method for detecting a lane crossing point according to an embodiment of the present application. It should be appreciated that the method may be applied to the scenario shown in fig. 1, and in particular to the cloud server 105 in fig. 1.

Illustratively, as shown in FIG. 2, the method 200 includes:

201, at least one first data point corresponding to a first lane line of a first road segment and a plurality of second data points corresponding to a second lane line of the first road segment are obtained from map data, the second lane line is a lane line adjacent to the first lane line, and when the first lane line is parallel to the second lane line, the running directions of the first lane line and the second lane line are the same.

It should be appreciated that in the crowd source map construction process, the cloud server may obtain map data required to construct the crowd source map by receiving road data collected by a plurality of vehicles.

Alternatively, the map data may include lane line data, building data, traffic light data, and the like, depending on the acquisition object.

In one possible implementation, the cloud server may process lane line data in the map data when lane line detection is required. In the embodiment of the application, the lane line data processing specifically refers to detecting whether the intersection point exists between two lane lines. Lane line data is understood to mean, in particular, a plurality of data points which form a lane line.

Since the position of each lane line on the road is fixed. Thus, lane line data acquired by the vehicle is also fixed. In other words, each data point of each lane line corresponds to a unique identification, which can be understood as coordinates of the data point, namely world geodetic system-84 (World Geodical System-84, wgs-84, also referred to as world geodetic system 84) coordinates based on the global positioning system (Global Positioning System, GPS). For convenience, in the embodiment of the present application, the seat of any one data point is labeled as "(x, y)", where x represents the longitude of the data point and y represents the latitude of the data point.

In the lane line detection process, the cloud server can judge whether two adjacent lane lines with the same driving direction exist or not based on lane line data of a plurality of lane lines acquired in advance. The reason why the traveling directions of two adjacent lane lines are defined herein to be the same is that there is no possibility that the two lane lines cross when the traveling directions of the two lane lines are different. Specifically, the driving directions of the two lane lines are the same, which means that when the two lane lines are parallel, the driving directions of the two lane lines are the same.

For example, for judging whether the driving directions of the two lane lines are the same, in the process of collecting the lane line data, the vehicle may send the driving direction of each lane line to the cloud server, so that when the cloud server detects the lane lines, it may determine whether the driving directions of the two adjacent lane lines are the same based on the driving directions.

When the driving directions of the two lane lines are determined to be the same, the cloud server can select one lane line (marked as a first lane line) as a current target lane line to process. Specifically, in the processing process, as the lane line corresponds to a plurality of data points, the cloud server traverses all data points corresponding to the lane line one by one along the driving direction of the lane line from the first data point. In traversing each data point, searching a plurality of data points of adjacent lane lines (marked as 'second lane lines') in a preset range by taking the current data point as a circle center, wherein the plurality of data points fall in the preset range.

Finally, the cloud server may jointly determine whether the first lane line and the second lane line intersect, and in the case of intersection, the intersection of the first lane line and the second lane line, based on the data points on the first lane line and the data points on the second lane line.

It should be appreciated that lane lines may correspond to either a merge or a merge when intersecting.

For the case where the lane lines merge, two lane lines on the road are generally almost parallel and the vertical distance between the two lane lines is kept within a certain range before the lane lines do not merge. When two lane lines cross, the vertical distance between the two lane lines can be suddenly changed at a certain point and gradually reduced until the two lane lines cross.

For the case where the lane lines merge, the two lane lines are almost coincident and the vertical distance between the two lane lines is substantially 0 before the lane lines do not merge. When two lane lines approach or reach the intersection point, the vertical distance between the two lane lines will be suddenly changed at a certain point and gradually increased until the two lane lines are gradually parallel.

The data acquisition process in both crossover cases is described in detail below with reference to fig. 3.

Fig. 3 is a schematic view of a scene of acquiring data points according to an embodiment of the present application.

Illustratively, as shown in (a) of fig. 3, a corresponding acquisition scenario of data points at the time of lane line entry is provided. The traveling direction is assumed to be from right to left. In the process of processing the lane line L1, the cloud server searches for a point 2, a point 3 and a point 4 on the lane line L2 in a certain range by taking the point 1 as a circle center in the searching process, wherein the data point which is processed currently is the point 1.

As shown in fig. 3 (b), the data point acquisition scenario corresponds to the data point acquisition scenario when the lane line is merged. Assume that the traveling direction is from left to right.

It should be understood that in the acquisition process of the embodiment of the present application, in order to ensure the accuracy of data acquisition of each lane line, each lane line is generally acquired repeatedly, and the data of the same lane line acquired repeatedly are fused, so as to finally obtain map data (lane line data) for constructing a crowd source map.

Alternatively, the lane line collecting mode may collect multiple data points on each lane line one by one according to the lane line sequence, or may collect multiple data points on each lane line in multiple lane lines at the same time.

It should also be understood that, although two lane lines intersect, in the actual collection process, the intersecting portion belongs to both the L1 lane line and the L2 lane line, that is, when the vehicle collects the L1 lane line, the data points of the intersecting portion can be obtained; data points of the intersection can also be acquired when the L2 lane line is acquired. In the process of processing the lane line L1, the cloud server assumes that the data point which is currently processed is a point 5, and in the process of searching, a point 6, a point 7 and a point 8 on the lane line L2 are searched in a certain range with the point 5 as a circle center.

The cloud server can acquire a plurality of data points on two lane lines within a distance in a mode shown in (a) - (b) in fig. 3, calculate the vertical distance between the two lane lines according to the plurality of data points, and judge whether the vertical distance between the two lane lines within the distance is suddenly changed or not so as to judge whether the two lane lines are possibly crossed or not.

Whether the vertical distance is abrupt or not, for the case of convergence, it means that the vertical distance is suddenly reduced; the term "convergence" refers to a sudden increase in vertical distance.

The distance may be a section of road where the currently processed data point is located, and is referred to as a "first section" in the embodiment of the present application, where the direction of the first section of road is the same as the traveling direction of the first lane line or the second lane line. When data points are acquired on a first road segment, the data points acquired on a first road segment are referred to as "at least one first data point" and the data points on a second road segment are referred to as "a plurality of second data points".

Through the step 201, the cloud server may obtain at least one data point on the first lane and a plurality of second data points on the second lane.

202, Determining whether at least one candidate intersection point exists between the first lane line and the second lane line according to the at least one first data point and the plurality of second data points, wherein the intersection probability between the first lane line and the second lane line is larger than the preset probability at the candidate intersection point.

Abrupt change of the vertical distance between the first lane line and the second lane line is a precondition that the two lane lines cross. Thus, after deriving the at least one first data point and the plurality of second data points, the cloud server may calculate a plurality of vertical distances between the first lane line and the second lane line based on the data points, determine whether a distance abrupt change occurs, and further determine whether at least one candidate intersection point exists between the first lane line and the second lane line.

In one possible implementation, determining whether at least one candidate intersection exists between the first lane line and the second lane line based on the at least one first data point and the plurality of second data points includes:

Judging whether the change rate of the vertical distance between the first lane line and the second lane line is larger than a preset change rate or not according to at least one first data point and a plurality of second data points;

When the change rate of the vertical distance is larger than the preset change rate, starting from the end position of the first road section, acquiring a plurality of third data points corresponding to the first lane line of the second road section and a plurality of fourth data points corresponding to the second lane line of the second road section from the map data, wherein the direction of the second road section is the same as the running direction;

Based on the third plurality of data points and the fourth plurality of data points, it is determined whether at least one candidate intersection exists between the first lane line and the second lane line.

Illustratively, as shown in (a) of fig. 3, at least one first data point is point 1 and a plurality of second data points are points 2,3 and 4. As shown in fig. 3 (b), at least one first data point is point 5, and a plurality of second data points are points 6, 7, and 8.

In calculating the vertical distance between the lane line L1 and the lane line L2, since L2 in fig. 3 is approximately a straight line, it can be calculated using the following point-to-straight line distance formula (1):

Formula (1)

Wherein, in formula (1):

Ax₀+By₀ +c: the equation of the lane line L2 can be obtained by solving any two data points in a plurality of second data points;

x₀: a longitude of each of the at least one first data point;

y₀: a latitude of each of the at least one first data point;

d: perpendicular distance between lane line L1 and lane line L2, unit: rice (m).

Through the mode, the cloud server can calculate a plurality of vertical distances between the first lane line and the second lane line in the first road section, and the plurality of vertical distances are assumed to be'd₁、d₂、d₃ and d₄' respectively.

In the case where the first lane line and the second lane line are parallel or coincide, the rate of change of the vertical distance is generally not too large. The vertical distance change rate can be expressed specifically herein as the following formula (2):

Wherein, in formula (2):

s: a vertical distance change rate;

d₁: the vertical distance calculated last time, unit: rice (m);

d₂: currently calculated vertical distance, unit: rice (m).

In general, for the case where the lane lines merge, the two lane lines are almost parallel before the lane lines do not merge, and the vertical distance between the lane lines is 3.5 to 3.7 meters. Based on this, the range of the change rate of the vertical distance is about 0 to 5%.

For the case of a lane departure, the two lane lines almost coincide before departure, the perpendicular distance between the lane lines being approximately 0 meters. Based on this, the range of the change rate of the vertical distance is about 0 to 5%. Therefore, the preset rate of change may be 5%. The above-mentioned preset change rate is only illustrative, and can be adjusted according to actual requirements in the actual application process.

After obtaining a plurality of vertical distance change rates corresponding to two adjacent vertical distances based on the plurality of vertical distances, the cloud server may determine whether the plurality of vertical distance change rates have a vertical distance change rate greater than a preset change rate.

When the vertical distance change rate is smaller than or equal to the preset change rate, the two lane lines are relatively parallel or coincide. When the vertical distance change rate is greater than the preset change rate, the curvature (or trajectory) of the lane line is suddenly changed, and there is a possibility of crossing.

Therefore, when the cloud server judges that the vertical distance change rate is greater than the preset change rate, the cloud server starts to acquire a plurality of third data points of the second road section on the first road section and a plurality of fourth data points of the second road section on the second road section along the driving direction from the end position of the first road section.

Alternatively, the length of the second road section and the length of the first road section may be the same or different, which is not limited in the embodiment of the present application.

When the third data points and the fourth data points in the second road section are acquired, the same manner as the aforementioned method for acquiring the at least one first data point and the plurality of second data points is not repeated here.

After deriving the third plurality of data points and the fourth plurality of data points, the cloud server may determine whether at least one candidate intersection exists between the first lane line and the second lane line.

Specifically, after the abrupt change of the vertical distance between the first lane line and the second lane line, when determining whether at least one candidate intersection point exists between the two lane lines according to the plurality of third data points and the plurality of fourth data points, it may be judged by the vertical distance between the two lane lines as well.

In one possible implementation, determining whether at least one candidate intersection exists between the first lane line and the second lane line based on the plurality of third data points and the plurality of fourth data points includes:

Determining a plurality of first vertical distances of the first lane line and the second lane line from the plurality of third data points and the plurality of fourth data points;

Determining a distance change trend according to the plurality of first vertical distances, wherein the distance change trend is used for representing the change rule of the vertical distances along the driving direction;

determining whether at least one candidate intersection exists between the first lane line and the second lane line according to the distance variation trend and/or the plurality of first vertical distances.

Fig. 4 is a schematic view of a scenario for calculating a vertical distance according to an embodiment of the present application.

For example, as shown in fig. 4 (a), the lane line is merged. Assume that the plurality of third data points are point a, point b, point c, and point d, respectively. The fourth plurality of data points are point e, point f, point g, point h, point i, and point j, respectively.

Through the foregoing formula (1), the cloud server can calculate the vertical distance d₁₄ of the point a on the lane line L1 from the lane line L2, the vertical distance d₁₃ of the point b on the lane line L1 from the lane line L2, the vertical distance d₁₂ of the point c on the lane line L1 from the lane line L2, and the vertical distance d₁₁ of the point d on the lane line L1 from the lane line L2, i.e. the plurality of first vertical distances.

After obtaining the plurality of first vertical distances, the cloud server may determine a distance change trend, that is, a change rule of the vertical distance along the driving direction, that is, whether the vertical distance continuously increases or continuously decreases along the driving direction, or whether the vertical distance does not exhibit a regular change, according to the plurality of first vertical distances.

Illustratively, as shown in fig. 4 (a), according to the magnitudes of the plurality of first vertical distances d₁₄、d₁₃、d₁₂ and d₁₁, the cloud server may determine that the distance change trend is decreasing.

As shown in fig. 4 (b), according to the magnitudes of the plurality of first vertical distances d₁₄、d₁₃、d₁₂ and d₁₁, the cloud server can determine that the distance change trend is increasing.

After determining the distance change trend, the cloud server can determine whether at least one candidate intersection point exists between the first lane line and the second lane line according to the distance change trend and/or the plurality of first vertical distances.

It should be appreciated that the reason why the plurality of first vertical distances are further considered in addition to the distance variation trend in determining the at least one candidate intersection point is that: when the distance change trend meets the crossing condition, if the vertical distance between the two lane lines is still large, it is indicated that the two lane lines are not close to crossing (or overlap), i.e. no candidate crossing point exists yet. Therefore, when the distance variation trend satisfies the crossing condition, it is also necessary to judge whether or not the two lane lines have crossed according to the plurality of first vertical distances.

When determining whether at least one candidate intersection exists between the first lane line and the second lane line according to the distance variation trend and/or the plurality of first vertical distances, on the one hand, it is required to determine whether the distance variation trend is continuously increasing or continuously decreasing, and on the other hand, it is also required to determine whether a first vertical distance smaller than or equal to a first threshold value exists among the plurality of first vertical distances, and the first threshold value may be regarded as a vertical distance representing when the first lane line and the second lane line are close to the intersection.

In a possible implementation manner, determining whether at least one candidate intersection exists between the first lane line and the second lane line according to the distance variation trend and/or the plurality of first vertical distances includes:

determining that at least one candidate intersection exists between the first lane line and the second lane line in the case where the distance variation trend satisfies a preset trend and at least one first vertical distance less than or equal to a first threshold exists among the plurality of first vertical distances;

and determining that at least one candidate intersection point does not exist between the first lane line and the second lane line under the condition that the distance change trend does not meet the preset trend or the plurality of first vertical distances are all larger than a first threshold value.

Optionally, the preset trend includes increasing or decreasing.

When the distance change trend increases or decreases, it is indicated that the two lane lines satisfy the intersection condition, i.e., merge or merge. Further, when there is at least one first vertical distance of the plurality of first vertical distances that is less than or equal to the first threshold value, it is indicated that there is a data point that is close to the intersection point, i.e., the current two lane lines are about to coincide, among the selected plurality of third data points and the plurality of fourth data points. The cloud server determines that there is at least one candidate intersection for the first lane line and the second lane line.

In contrast, when the distance change trend is not increased or decreased, the sink or sink condition is not satisfied. Or when the plurality of first vertical distances are each greater than the first threshold, indicating that the selected data point is not already a data point near the intersection, the cloud server determining that at least one candidate intersection is not present between the first lane line and the second lane line.

In another scenario, on the basis of the vertical distance, it may further combine the slope change trend of the lane line to determine whether at least one candidate intersection exists in the lane line.

For example, as shown in (a) and (b) in fig. 4, there are a plurality of data points on the lane line L1, and the points a, b, c and d may be calculated by the cloud server according to the coordinates of the points a, b, c and d in addition to the points d₁₄、d₁₃、d₁₂ and d₁₁, so as to obtain three slopes of the lane line L1 along the driving direction, which are respectively denoted as "k_cd、k_bc and k_ab". When d₁₄＜d₁₃＜d₁₂＜d₁₁ and d₁₄ are less than or equal to the first threshold, and k_cd＜k_bc＜k_ab, then it is indicated that there is a candidate intersection between lane line L1 and lane line L2.

After determining that at least one candidate intersection exists between the first lane line and the second lane line, the cloud server may further determine the intersection type of the two lane lines, i.e. whether the first lane line and the second lane line are specifically a lane line junction or a lane line junction.

In one possible implementation, after determining that there is at least one candidate intersection between the first lane line and the second lane line, the method further includes:

And determining the crossing type of the first lane line and the second lane line according to the distance change trend, wherein the crossing type is used for indicating the crossing direction of the first lane line and the second lane line.

And under the condition that the distance change trend accords with the preset trend, the distance change trend is either increased or decreased. Based on the two distance change trends, two crossing types are corresponding.

In a possible implementation manner, determining a crossing type of the first lane line and the second lane line according to a distance variation trend includes:

under the condition that the distance change trend is reduced, determining that the intersection type is lane line convergence;

in the case where the distance change trend is increasing, the intersection type is determined as the lane line junction.

Illustratively, as shown in fig. 4 (a), according to the magnitudes of the plurality of first vertical distances d₁₄、d₁₃、d₁₂ and d₁₁, the cloud server may determine that the distance change trend is decreasing. Thus, the intersection type is lane line merging.

As shown in fig. 4 (b), according to the magnitudes of the plurality of first vertical distances d₁₄、d₁₃、d₁₂ and d₁₁, the cloud server can determine that the distance change trend is increasing. Thus, the intersection type is lane line junction.

After determining that at least one candidate intersection exists between the first lane line and the second lane line, the cloud server may determine the at least one candidate intersection based on the plurality of first vertical distances.

In one possible implementation, after determining that at least one candidate intersection exists between the first lane line and the second lane line, the method further includes:

at least one candidate intersection is determined as at least one first data point of the plurality of first data points corresponding to the at least one first vertical distance and at least one second data point of the plurality of second data points corresponding to the at least one first vertical distance.

Illustratively, as shown in fig. 4 (a) and (b), the plurality of first vertical distances are d₁₄、d₁₃、d₁₂ and d₁₁, respectively, assuming d₁₄ is less than the first threshold. If the third data point used in calculating d₁₄ is the point a and the fourth data points e and f, if the cloud server determines that d₁₄ is less than the first threshold, the data points for calculating d₁₄, namely, the point a, the point e and the point f, can be determined, that is, at least one candidate intersection meeting the condition.

Thus, through step 202, in the case of determining that the first lane line and the second lane line intersect, the cloud server may further determine at least one candidate intersection point that meets the condition.

203, In case there is at least one candidate intersection between the first lane line and the second lane line, determining a target intersection between the first lane line and the second lane line based on the at least one candidate intersection, the target intersection being used to represent the actual intersection position of the first lane line and the second lane line.

After determining the at least one candidate intersection, the cloud server may also process the at least one candidate intersection in order to obtain a more accurate target intersection.

In a possible implementation manner, in a case where at least one candidate intersection exists between the first lane line and the second lane line, determining a target intersection between the first lane line and the second lane line according to the at least one candidate intersection includes:

fusing at least one candidate intersection to obtain a predicted intersection;

when the distance change trend is reduced, starting with the predicted intersection point, acquiring a plurality of fifth data points corresponding to a first lane line of a third road section and a plurality of sixth data points corresponding to a second lane line of the third road section, wherein the direction of the third road section is the same as the running direction; determining a target intersection from the predicted intersection, the plurality of fifth data points, and the plurality of sixth data points;

when the distance change trend is increased, starting with a predicted intersection point, acquiring a plurality of seventh data points corresponding to a first lane line of a fourth road section and a plurality of eighth data points corresponding to a second lane line of the fourth road section, wherein the direction of the fourth road section is opposite to the driving direction; a target intersection is determined based on the predicted intersection, the seventh plurality of data points, and the eighth plurality of data points.

Illustratively, as shown in fig. 4 (a) and (b), at least one candidate intersection is point a, point e, and point f, since each data point corresponds to coordinates at the time of acquisition. Thus, based on the coordinates of the at least one candidate intersection, the cloud server may fuse the at least one candidate intersection, i.e., calculate an average value (average coordinate) of the coordinates of the at least one candidate intersection, and determine a data point corresponding to the average coordinate as the predicted intersection.

In order to ensure the accuracy and reliability of the intersection judgment of the lane lines, after the predicted intersection point is obtained, the cloud server can also use the predicted intersection point as a starting point to obtain the vertical distance between two lane lines within a certain distance after the intersection point, and judge whether the predicted intersection point can be used as a final required target intersection point.

It should be understood that when the perpendicular distance of two lane lines within a distance after the intersection is acquired with the predicted intersection as the starting point, for the case of the lane line junction, in the process of traversing the data points on the lane line in the traveling direction of the lane line, since the data points within a distance after the intersection in this case have been traversed, the cloud server needs to calculate the perpendicular distance again in order to secure the stability of the intersection.

For example, as shown in fig. 4 (a), in the case where the lane lines merge, assuming that the determined predicted intersection is the point O, the cloud server may acquire, along the driving direction, a plurality of fifth data points (not shown in the drawing) corresponding to the first lane line in the third path section and a plurality of sixth data points (not shown in the drawing) corresponding to the second lane line in the third path section according to the data acquisition manner shown in fig. 3.

Based on the point O, the plurality of fifth data points, and the plurality of sixth data points, the cloud server may determine whether the predicted intersection can be the target intersection if the lane lines merge.

Specifically, when judging whether the predicted intersection point can be the target intersection point, the cloud server may continue to be implemented by calculating the vertical distance corresponding to the formula (1).

In one possible implementation, the method further includes determining a target intersection from the predicted intersection, the plurality of fifth data points, and the plurality of sixth data points:

Determining a plurality of second vertical distances from the plurality of fifth data points and the plurality of sixth data points; and determining the target intersection as a predicted intersection in the case that the plurality of second vertical distances are all less than or equal to the second threshold.

Specifically, after obtaining the fifth data points and the sixth data points, the cloud server calculates a plurality of second vertical distances between the first lane line and the second lane line according to formula (1). When the plurality of second vertical distances are each less than or equal to the second threshold value, it is indicated that the first lane line and the second lane line remain stably coincident after the predicted intersection, in which case the predicted intersection may be determined as the final target intersection.

Based on the point O, the plurality of seventh data points, and the plurality of eighth data points, the cloud server may determine whether the predicted intersection can be the target intersection if the lane lines merge.

In one possible implementation, determining the target intersection from the predicted intersection, the seventh plurality of data points, and the eighth plurality of data points includes:

Determining a plurality of third vertical distances from the seventh plurality of data points and the eighth plurality of data points; and determining the target intersection as a predicted intersection in the case that the plurality of third vertical distances are all less than or equal to the second threshold.

Specifically, after obtaining a plurality of seventh data points and a plurality of eighth data points, the cloud server calculates a plurality of third vertical distances between the first lane line and the second lane line according to formula (1). When the plurality of third vertical distances are each less than or equal to the second threshold value, it is indicated that the first lane line and the second lane line remain stably coincident after the predicted intersection, in which case the predicted intersection may be determined as the final target intersection.

When two lane lines are converged, after a predicted intersection point is obtained, the method can further determine a plurality of second vertical distances in a third path segment after the predicted intersection point according to a plurality of fifth data points and a plurality of sixth data points, and when the plurality of second vertical distances are smaller than or equal to a second threshold value, the predicted intersection point is determined to be a target intersection point. Similarly, when two lane lines are converged, a plurality of third vertical distances in the fourth road section after the predicted intersection point can be determined according to a plurality of seventh data points and a plurality of eighth data points, and when the plurality of third vertical distances are smaller than or equal to a second threshold value, the process of determining the predicted intersection point as the target intersection point can ensure the stability of the intersection of the lane lines and avoid inaccurate intersection point determination caused by the fact that only local areas of the two lane lines are intersected.

Fig. 5 is a schematic structural diagram of a device for detecting a lane crossing according to an embodiment of the present application.

Illustratively, as shown in FIG. 5, the apparatus 500 includes:

The data obtaining module 501 is configured to obtain, from map data, at least one first data point corresponding to a first lane line of a first road segment and a plurality of second data points corresponding to a second lane line of the first road segment, where the second lane line is a lane line adjacent to the first lane line, and when the first lane line is parallel to the second lane line, a driving direction of the first lane line and a driving direction of the second lane line are the same;

A candidate intersection determination module 502, configured to determine, according to the at least one first data point and the plurality of second data points, whether at least one candidate intersection exists between the first lane line and the second lane line, where a probability of intersection between the first lane line and the second lane line is greater than a preset probability;

A target intersection determination module 503, configured to determine a target intersection between the first lane line and the second lane line according to the at least one candidate intersection when the at least one candidate intersection exists between the first lane line and the second lane line, where the target intersection is used to represent an actual intersection position of the first lane line and the second lane line.

In a possible implementation manner, the candidate intersection judgment module 502 is specifically configured to: judging whether the change rate of the vertical distance between the first lane line and the second lane line is greater than a preset change rate according to the at least one first data point and the plurality of second data points; when the change rate of the vertical distance is larger than the preset change rate, starting from the end position of the first road section, acquiring a plurality of third data points corresponding to the first road line of a second road section and a plurality of fourth data points corresponding to the second road line of the second road section from the map data, wherein the direction of the second road section is the same as the driving direction; determining whether the at least one candidate intersection exists between the first lane line and the second lane line based on the plurality of third data points and the plurality of fourth data points.

In a possible implementation manner, the candidate intersection judgment module 502 is further configured to: determining a plurality of first vertical distances of the first lane line and the second lane line from the plurality of third data points and the plurality of fourth data points; determining a distance change trend according to the plurality of first vertical distances, wherein the distance change trend is used for representing the change rule of the vertical distances along the driving direction; determining whether the at least one candidate intersection exists between the first lane line and the second lane line according to the distance variation trend and/or the plurality of first vertical distances.

In a possible implementation manner, the candidate intersection judgment module 502 is further configured to: determining that the at least one candidate intersection exists between the first lane line and the second lane line in the case where the distance variation trend satisfies a preset trend and at least one first vertical distance less than or equal to a first threshold exists among the plurality of first vertical distances; and determining that the at least one candidate intersection point does not exist between the first lane line and the second lane line in the case that the distance variation trend does not satisfy the preset trend or the plurality of first vertical distances are all greater than the first threshold.

Optionally, after the determining that the at least one candidate intersection exists between the first lane line and the second lane line, the apparatus further comprises: a candidate intersection determination module for determining the at least one candidate intersection as at least one third data point of the plurality of third data points corresponding to the at least one first vertical distance and at least one fourth data point of the plurality of fourth data points corresponding to the at least one first vertical distance.

In a possible implementation manner, the preset trend includes increasing or decreasing, and the target intersection determining module 503 is specifically configured to: fusing the at least one candidate intersection to obtain a predicted intersection; when the distance change trend is reduced, starting with the predicted intersection point, acquiring a plurality of fifth data points corresponding to the first road line of a third road section and a plurality of sixth data points corresponding to the second road line of the third road section, wherein the direction of the third road section is the same as the driving direction; determining the target intersection based on the predicted intersection, the plurality of fifth data points, and the plurality of sixth data points; when the distance change trend is increased, starting with the predicted intersection point, acquiring a plurality of seventh data points corresponding to the first lane line of a fourth road section and a plurality of eighth data points corresponding to the second lane line of the fourth road section, wherein the direction of the fourth road section is opposite to the driving direction; the target intersection is determined based on the predicted intersection, the plurality of seventh data points, and the plurality of eighth data points.

In a possible implementation, the target intersection determination module 503 is further configured to: determining a plurality of second vertical distances from the plurality of fifth data points and the plurality of sixth data points; determining the target intersection as the predicted intersection if the plurality of second vertical distances are each less than or equal to a second threshold; and determining a plurality of third vertical distances from the plurality of seventh data points and the plurality of eighth data points; and determining the target intersection as the predicted intersection when the plurality of third vertical distances are all less than or equal to the second threshold.

Optionally, after the determining that the at least one candidate intersection exists between the first lane line and the second lane line, the apparatus further comprises: and the intersection type determining module is used for determining the intersection type of the first lane line and the second lane line according to the distance change trend, wherein the intersection type is used for representing the intersection direction of the first lane line and the second lane line.

In a possible implementation manner, the preset trend includes increasing or decreasing, the intersection type includes lane line in or lane line out, and the intersection type determining module is specifically configured to: determining that the intersection type is lane line merging under the condition that the distance change trend is reduced; in the case where the distance change trend is increasing, the intersection type is determined as a lane line junction.

Illustratively, as shown in fig. 6, the cloud server 105 includes: a memory 601 and a processor 602, wherein the memory 601 stores executable program code 6011, and the processor 602 is configured to invoke and execute the executable program code 6011 to perform a method of detecting a lane crossing.

In this embodiment, the cloud server may be divided into functional modules according to the above method example, for example, each functional module may be corresponding to one processing module, or two or more functions may be integrated into one processing module, where the integrated modules may be implemented in a hardware form. It should be noted that, in this embodiment, the division of the modules is schematic, only one logic function is divided, and another division manner may be implemented in actual implementation.

In the case of dividing each function module with corresponding each function, the cloud server may include: a data acquisition module, a candidate intersection judgment module, a target intersection determination module and the like. It should be noted that, all relevant contents of each step related to the above method embodiment may be cited to the functional descriptions of the corresponding functional modules, which are not described herein.

The cloud server provided in the embodiment is used for executing the method for detecting the lane line intersection, so that the same effect as that of the implementation method can be achieved.

In case of an integrated unit, the cloud server may comprise a processing module, a storage module. The processing module can be used for controlling and managing actions of the cloud server. The storage module may be used to support the cloud server to execute relevant program code, data, and the like.

Wherein the processing module may be a processor or controller that may implement or execute the various illustrative logical blocks, modules, and circuits described in connection with the present disclosure. A processor may also be a combination of computing functions, including for example one or more microprocessors, digital Signal Processing (DSP) and microprocessor combinations, etc., and a memory module may be a memory.

The present embodiment also provides a computer-readable storage medium having stored therein computer program code which, when run on a computer, causes the computer to perform the above-described related method steps to implement a method of detecting lane-line intersections in the above-described embodiments.

The present embodiment also provides a computer program product which, when run on a computer, causes the computer to perform the above-described related steps to implement a method of detecting lane-line intersections in the above-described embodiments.

In addition, the cloud server provided by the embodiment of the application can be a chip, a component or a module, and the cloud server can comprise a processor and a memory which are connected; the memory is used for storing instructions, and when the cloud server runs, the processor can call and execute the instructions to enable the chip to execute the method for detecting the lane line crossing point in the embodiment.

The cloud server, the computer readable storage medium, the computer program product or the chip provided in this embodiment are all configured to execute the corresponding method provided above, so that the beneficial effects thereof can be referred to the beneficial effects in the corresponding method provided above, and will not be described herein.

It will be appreciated by those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional modules is illustrated, and in practical application, the above-described functional allocation may be performed by different functional modules according to needs, i.e. the internal structure of the apparatus is divided into different functional modules to perform all or part of the functions described above.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of modules or units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another apparatus, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the application is subject to the protection scope of the claims.

Claims

1. A method of detecting lane-crossing points, the method comprising:

Acquiring at least one first data point corresponding to a first lane line of a first road segment and a plurality of second data points corresponding to a second lane line of the first road segment from map data, wherein the second lane line is a lane line adjacent to the first lane line, and when the first lane line is parallel to the second lane line, the running directions of the first lane line and the second lane line are the same;

Determining whether at least one candidate intersection point exists between the first lane line and the second lane line according to the at least one first data point and the plurality of second data points, wherein the intersection probability between the first lane line and the second lane line is larger than a preset probability at the candidate intersection point;

And in the case that the at least one candidate intersection point exists between the first lane line and the second lane line, determining a target intersection point between the first lane line and the second lane line according to the at least one candidate intersection point, wherein the target intersection point is used for representing the actual intersection position of the first lane line and the second lane line.

2. The method of claim 1, wherein the determining whether at least one candidate intersection exists between the first lane line and the second lane line based on the at least one first data point and the plurality of second data points comprises:

Judging whether the change rate of the vertical distance between the first lane line and the second lane line is larger than a preset change rate according to the at least one first data point and the plurality of second data points;

When the change rate of the vertical distance is larger than the preset change rate, starting from the end position of the first road section, acquiring a plurality of third data points corresponding to the first lane line of a second road section and a plurality of fourth data points corresponding to the second lane line of the second road section from the map data, wherein the direction of the second road section is the same as the running direction;

Determining whether the at least one candidate intersection exists between the first lane line and the second lane line based on the plurality of third data points and the plurality of fourth data points.

3. The method of claim 2, wherein the determining whether the at least one candidate intersection exists between the first lane line and the second lane line based on the plurality of third data points and the plurality of fourth data points comprises:

Determining a distance change trend according to the plurality of first vertical distances, wherein the distance change trend is used for representing a change rule of the vertical distances along the running direction;

And determining whether the at least one candidate intersection exists between the first lane line and the second lane line according to the distance change trend and/or the plurality of first vertical distances.

4. A method according to claim 3, wherein said determining whether said at least one candidate intersection exists between said first lane-line and said second lane-line based on said distance trend and/or said plurality of first vertical distances comprises:

Determining that there is the at least one candidate intersection between the first lane line and the second lane line in a case where the distance variation trend satisfies a preset trend and at least one first vertical distance that is less than or equal to a first threshold exists among the plurality of first vertical distances;

and determining that the at least one candidate intersection point does not exist between the first lane line and the second lane line in the case that the distance variation trend does not satisfy the preset trend or the plurality of first vertical distances are all greater than the first threshold value.

5. The method of claim 4, wherein after the determining that the at least one candidate intersection exists between the first lane line and the second lane line, the method further comprises:

determining the at least one candidate intersection as at least one third data point of the plurality of third data points corresponding to the at least one first vertical distance and at least one fourth data point of the plurality of fourth data points corresponding to the at least one first vertical distance.

6. The method of claim 5, wherein the predetermined trend includes increasing or decreasing, and wherein the determining a target intersection between the first lane line and the second lane line based on the at least one candidate intersection if the at least one candidate intersection exists between the first lane line and the second lane line comprises:

Fusing the at least one candidate intersection point to obtain a predicted intersection point;

When the distance change trend is reduced, starting with the prediction intersection, acquiring a plurality of fifth data points corresponding to the first lane line of a third road section and a plurality of sixth data points corresponding to the second lane line of the third road section, wherein the direction of the third road section is the same as the driving direction; determining the target intersection from the predicted intersection, the plurality of fifth data points, and the plurality of sixth data points;

when the distance change trend is increased, starting with the predicted intersection point, acquiring a plurality of seventh data points corresponding to the first lane line of a fourth road section and a plurality of eighth data points corresponding to the second lane line of the fourth road section, wherein the direction of the fourth road section is opposite to the driving direction; the target intersection is determined from the predicted intersection, the plurality of seventh data points, and the plurality of eighth data points.

7. The method of claim 6, wherein the determining the target intersection from the predicted intersection, the plurality of fifth data points, and the plurality of sixth data points comprises:

determining a plurality of second vertical distances from the plurality of fifth data points and the plurality of sixth data points; determining the target intersection as the predicted intersection if the plurality of second vertical distances are each less than or equal to a second threshold;

And said determining said target intersection from said predicted intersection, said plurality of seventh data points, and said plurality of eighth data points, comprising:

Determining a plurality of third vertical distances from the plurality of seventh data points and the plurality of eighth data points; and determining the target intersection as the predicted intersection when the plurality of third vertical distances are all less than or equal to the second threshold.

8. The method of claim 4, wherein after the determining that the at least one candidate intersection exists between the first lane line and the second lane line, the method further comprises:

and determining the crossing type of the first lane line and the second lane line according to the distance change trend, wherein the crossing type is used for representing the crossing direction of the first lane line and the second lane line.

9. The method of claim 8, wherein the predetermined trend includes increasing or decreasing, the intersection type includes lane-line-in or lane-line-out, and the determining the intersection type of the first lane line and the second lane line according to the distance change trend includes:

Determining that the intersection type is lane line merging under the condition that the distance change trend is reduced;

And determining that the intersection type is a lane departure when the distance change trend is increased.

10. An apparatus for detecting a lane-crossing, the apparatus comprising:

The system comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring at least one first data point corresponding to a first lane line of a first road section and a plurality of second data points corresponding to a second lane line of the first road section from map data, the second lane line is a lane line adjacent to the first lane line, and when the first lane line is parallel to the second lane line, the running directions of the first lane line and the second lane line are the same;

A candidate intersection judgment module, configured to determine, according to the at least one first data point and the plurality of second data points, whether at least one candidate intersection exists between the first lane line and the second lane line, where a probability of intersection between the first lane line and the second lane line is greater than a preset probability;

a target intersection determination module, configured to determine, in a case where the at least one candidate intersection exists between the first lane line and the second lane line, a target intersection between the first lane line and the second lane line according to the at least one candidate intersection, where the target intersection is used to represent an actual intersection position of the first lane line and the second lane line.

11. A cloud server, the cloud server comprising:

A memory for storing executable program code;

A processor for calling and running the executable program code from the memory, causing the cloud server to perform the method of any of claims 1 to 9.

12. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program which, when executed, implements the method according to any one of claims 1 to 9.