CN116467615A

CN116467615A - Clustering method and device for vehicle tracks, storage medium and electronic device

Info

Publication number: CN116467615A
Application number: CN202310403878.9A
Authority: CN
Inventors: 徐佳; 韩茂强; 张星宇; 杜宇航; 王庆飞
Original assignee: Beijing Wanji Technology Co Ltd
Current assignee: Beijing Wanji Technology Co Ltd
Priority date: 2023-04-14
Filing date: 2023-04-14
Publication date: 2023-07-21

Abstract

The application discloses a clustering method and device for vehicle tracks, a storage medium and an electronic device, wherein the method comprises the following steps: according to the number N of the drivable routes of the target intersection, N reference tracks are selected from a group of vehicle tracks of the target intersection, wherein each vehicle track in the group of vehicle tracks is a section of driving track from the target intersection to the departure of a vehicle corresponding to each vehicle track, and N is a positive integer greater than or equal to 2; respectively selecting a group of clustering center points from each of N reference tracks to obtain initial N groups of clustering center points; and performing clustering operation on a group of vehicle tracks according to the initial N groups of clustering center points to obtain a clustering result of the group of vehicle tracks, wherein the clustering result of the group of vehicle tracks is used for indicating N track class clusters obtained by clustering the group of vehicle tracks and the vehicle tracks contained in each track class cluster in the N track class clusters.

Description

Clustering method and device for vehicle tracks, storage medium and electronic device

Technical Field

The application relates to the field of vehicle wireless communication, in particular to a clustering method and device for vehicle tracks, a storage medium and an electronic device.

Background

At present, in an uncertain and dynamic traffic environment, the characteristics of vehicle tracks at all intersections are known, similar tracks are clustered, hidden rules in the tracks are mined, and an effective data set can be provided for vehicle track prediction, vehicle behavior analysis and the like, so that driving guidance is provided for vehicles in time, and road traffic safety and traffic efficiency are improved.

In a general clustering method, the Euclidean distance between points to be classified and a clustering center is generally calculated directly, but for vehicle tracks, particularly complex intersections, the number of the vehicle running routes is large, the running direction and the running lane corresponding to each running route are different, and one clustering center point cannot accurately represent the position of the track, so that the track clustering accuracy is poor.

Therefore, the clustering method of the vehicle track in the related technology has the problem of poor clustering accuracy caused by complex intersection environments.

Disclosure of Invention

The embodiment of the application provides a vehicle track clustering method and device, a storage medium and an electronic device, which are used for at least solving the problem that the clustering accuracy of the vehicle track clustering method in the related technology is poor due to complex intersection environment.

According to an aspect of an embodiment of the present application, there is provided a clustering method of vehicle trajectories, including: selecting N reference tracks from a group of vehicle tracks of a target intersection according to the number N of drivable routes of the target intersection, wherein each vehicle track in the group of vehicle tracks is a section of driving track from the target intersection to the departure of a vehicle corresponding to each vehicle track, and N is a positive integer greater than or equal to 2; respectively selecting a group of clustering center points from each of the N reference tracks to obtain initial N groups of clustering center points; and performing clustering operation on the group of vehicle tracks according to the initial N groups of clustering center points to obtain a clustering result of the group of vehicle tracks, wherein the clustering result of the group of vehicle tracks is used for indicating N track class clusters obtained by clustering the group of vehicle tracks and vehicle tracks contained in each track class cluster in the N track class clusters.

According to another aspect of the embodiments of the present application, there is also provided a clustering device for vehicle tracks, including: a selection unit, configured to select N reference tracks from a set of vehicle tracks of a target intersection according to a number N of drivable routes of the target intersection, where each vehicle track in the set of vehicle tracks is a section of a driving track from the target intersection to a vehicle leaving the target intersection, where N is a positive integer greater than or equal to 2; the selecting unit is used for respectively selecting a group of clustering center points from each of the N reference tracks to obtain initial N groups of clustering center points; the execution unit is used for executing clustering operation on the group of vehicle tracks according to the initial N groups of clustering center points to obtain a clustering result of the group of vehicle tracks, wherein the clustering result of the group of vehicle tracks is used for indicating N track class clusters obtained by clustering the group of vehicle tracks and vehicle tracks contained in each track class cluster in the N track class clusters.

According to yet another aspect of the embodiments of the present application, there is also provided a computer readable storage medium having a computer program stored therein, wherein the computer program is configured to perform the above-described clustering method of vehicle trajectories when running.

According to still another aspect of the embodiments of the present application, there is further provided an electronic device including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the clustering method of vehicle trajectories described above through the computer program.

In the embodiment of the application, a clustering operation mode of vehicle tracks is adopted by using a group of clustering center points on the vehicle tracks, N reference tracks are selected from a group of vehicle tracks of a target intersection according to the number N of drivable routes of the target intersection, wherein each vehicle track in the group of vehicle tracks is a section of driving track of a vehicle corresponding to each vehicle track from the target intersection to the departure from the target intersection, and N is a positive integer greater than or equal to 2; respectively selecting a group of clustering center points from each of N reference tracks to obtain initial N groups of clustering center points; according to the initial N groups of clustering center points, clustering operation is carried out on a group of vehicle tracks to obtain a group of clustering results of the vehicle tracks, wherein the clustering results of the group of the vehicle tracks are used for indicating N track class clusters obtained by clustering the group of the vehicle tracks and vehicle tracks contained in each track class cluster in the N track class clusters.

Drawings

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

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required to be used in the description of the embodiments or the prior art will be briefly described below, and it will be obvious to those skilled in the art that other drawings can be obtained from these drawings without inventive effort.

FIG. 1 is a schematic illustration of a hardware environment of an alternative clustering method of vehicle trajectories according to an embodiment of the present application;

FIG. 2 is a flow chart of an alternative clustering method of vehicle trajectories according to an embodiment of the present application;

FIG. 3 is a schematic illustration of an alternative clustering method of vehicle trajectories according to an embodiment of the present application;

FIG. 4 is a block diagram of an alternative vehicle track clustering device in accordance with an embodiment of the present application;

fig. 5 is a block diagram of an alternative electronic device according to an embodiment of the present application.

Detailed Description

In order to make the present application solution better understood by those skilled in the art, the following description will be made in detail and with reference to the accompanying drawings in the embodiments of the present application, it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.

It should be noted that the terms "first," "second," and the like in the description and claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that embodiments of the present application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

According to one aspect of the embodiments of the present application, a method for clustering vehicle trajectories is provided. Alternatively, in the present embodiment, the clustering method of vehicle trajectories described above may be applied to a hardware environment including the internet of vehicles device 102 and the server 104 as shown in fig. 1. As shown in fig. 1, the server 104 is connected to the internet of vehicle device 102 via a network, and a database may be provided on the server or independent of the server for providing data storage services for the server 104. Here, the internet of vehicle device 102 may include an on-board V2X device located on a vehicle.

The network may include, but is not limited to, at least one of: wired network, wireless network. The wired network may include, but is not limited to, at least one of: a wide area network, a metropolitan area network, a local area network, and the wireless network may include, but is not limited to, at least one of: WIFI (Wireless Fidelity ), bluetooth.

The clustering method of the vehicle tracks in the embodiment of the application may be executed by the server 104, or may be executed by the server 104 and the internet of vehicles device 102 together. Taking the example that the clustering method of the vehicle tracks in the present embodiment is executed by the server 104 as an example, fig. 2 is a schematic flow chart of an alternative clustering method of the vehicle tracks according to an embodiment of the present application, as shown in fig. 2, the flow of the method may include the following steps:

step S202, selecting N reference tracks from a group of vehicle tracks of the target intersection according to the number N of drivable routes of the target intersection, wherein each vehicle track in the group of vehicle tracks is a section of driving track from the target intersection to the departure of the vehicle corresponding to each vehicle track, and N is a positive integer greater than or equal to 2.

The clustering method of the vehicle tracks in the embodiment can be applied to a scene of clustering analysis of the vehicle tracks of the target intersection. Here, the target intersection may be an intersection such as an intersection, a herringbone intersection, or a three-way intersection. The lane layout of the target intersection is various and complex, and vehicles taking the same position of the target intersection as a starting point can have various driving route selections. The clustering analysis of the vehicle track of the target intersection may refer to track classification of the vehicle track of the target intersection, so as to obtain a track classification cluster. The number of the track classification clusters corresponds to the number of the drivable paths, and under different clustering requirements, the drivable paths can have different meanings, for example, the clustering requirements are that a group of vehicle tracks are clustered to obtain track clustering results of the drivable paths in all driving directions, wherein the driving directions are divided into driving directions indicated by left turning, straight running, right turning and the like; or clustering requirements are that a group of vehicle tracks are clustered to obtain track clustering results of the lane-level driving routes. It will be appreciated that in some intersection forms, the number of drivable paths obtained by the two meanings may be the same, for example, lanes and driving directions are in one-to-one correspondence; under other road junction forms, the number of the drivable paths obtained by the two road junction forms is generally different, for example, when a certain road section has a plurality of straight lines or a plurality of left-turn lanes and a plurality of right-turn lanes, the lanes in the same direction are combined according to the driving direction clustering, and the driving paths formed by the independent lanes are clustered independently according to the lane level clustering. Of course, the meaning of the drivable routes may also be other forms, and the number of drivable routes is related to the lane layout of the intersections, and the purpose/requirement of the clustering.

In an uncertain and dynamic traffic environment, the characteristics of the vehicle track are known, similar tracks are clustered, the characteristics of the clustered track clusters are extracted by using a Gaussian regression method and the like according to the track classification result, and the method can be used for predicting the vehicle track of the same target intersection or the intersection with the same type specification (such as the width of a lane of the intersection, the rule of the driving direction, the layout of the lane and the like), vehicle navigation, lane anchoring, collision prediction and monitoring and scheduling, an effective data set is provided for vehicle behavior analysis and the like, and meanwhile, the clustered track subsets are used for learning or application, so that the processing time and the prediction result of the functions of subsequent track prediction and the like can be effectively improved.

In this embodiment, N reference trajectories may be selected from a set of vehicle trajectories at the target intersection according to the number N of drivable paths at the target intersection. Here, each vehicle track in the set of vehicle tracks may be a travel track of a vehicle corresponding to each vehicle track from the target intersection to the departure from the target intersection, and N may be a positive integer greater than or equal to 2. N reference tracks can be used for determining an initial clustering center point, and can be selected from a group of vehicle tracks, can be randomly selected or can be selected according to characteristics such as a target intersection, the vehicle tracks and the like, and the embodiment is not limited to the method.

The N drivable paths of the target intersection may be determined according to the lane layout condition of the target intersection, the actual drivable directions of each lane may be different (straight, left-turn, right-turn, etc.), and the drivable directions of the plurality of lanes may be the same (straight, left-turn, right-turn, etc.). For a general traffic scene, such as a vehicle track condition at an intersection, the classification number k of the track clusters, that is, the number of the aforementioned N reference tracks, may be set according to the number of routes that can actually be driven. When the number k of the track clusters is determined, the actually studied target intersection is required to be determined, all routes which can be possibly driven by the vehicle actually at the target intersection are determined according to the traffic rule of the target intersection, the classification number k of the track clusters is determined by calculating the sum of the number of the driving routes which can be carried out in each lane leading to the intersection, the uncertainty of the solving process of the classification number can be reduced, and the influence on the track clustering effect is avoided.

For example, taking a bidirectional four-lane intersection as an example, assuming that the left lane in the same direction can be directly moved and turned left, and the right lane in the same direction can be directly moved and turned right, the clustering requirement is that the vehicle tracks are clustered according to the number of the drivable directions, if the drivable paths can be determined to be k= 4*4 according to the drivable directions of the lanes, and correspondingly, when the above-mentioned group of vehicle tracks is clustered, the classification number can be set to k= 4*4. It should be noted that the actual driving directions of different lanes at the same intersection are not necessarily the same, and the classification number k is determined only according to the actual condition of the road and the clustering requirement.

Step S204, a group of clustering center points are selected from each of the N reference tracks respectively, and initial N groups of clustering center points are obtained.

In general clustering methods, such as a K-means (i.e., K-means clustering algorithm) method, the euclidean distance between a point to be classified and a clustering center is usually calculated directly, but for vehicle tracks, especially complex intersections, the number of available running routes of vehicles is large, the corresponding running direction and running lane of each running route are different, and for example, the intersection is taken as an example, tracks in different directions such as straight running, left turning, right turning and the like exist in the intersection at the same time, and one clustering center point cannot accurately represent the position of the track, so that the clustering result is greatly different from the actual vehicle track.

In order to solve at least some of the above problems, in this embodiment, considering that the vehicle track at the target intersection has a certain rule, the vehicle tracks may be clustered according to the spatiotemporal similarity of the vehicle tracks. Namely, selecting a group of clustering center points according to the track characteristics of the vehicle tracks, and clustering each vehicle track.

In this embodiment, a set of cluster center points may be selected from each of the N reference tracks, to obtain an initial N set of cluster center points. Here, a set of cluster center points may include a plurality of cluster center points. The initial N sets of cluster centers may be used for the initial computation process in the clustering operation. Correspondingly, according to the selection mode of a group of clustering center points, a group of points to be classified can be determined in each vehicle track and used for carrying out related calculation of Euclidean distance with N groups of clustering center points.

Alternatively, the set of clustering center points may be effective feature points in each reference track, and compared with one clustering center point, track clustering is further performed according to the set of feature points, so that the clustered track clusters can be more effectively represented.

Step S206, clustering operation is performed on a group of vehicle tracks according to the initial N groups of clustering center points, so that a clustering result of the group of vehicle tracks is obtained, wherein the clustering result of the group of vehicle tracks is used for indicating N track class clusters obtained by clustering the group of vehicle tracks and vehicle tracks contained in each track class cluster in the N track class clusters.

After the initial N groups of clustering center points are determined, a clustering operation may be performed on a group of vehicle tracks according to the initial N groups of clustering center points, to obtain a clustering result of a group of vehicle tracks. Here, the clustering result of the set of vehicle tracks may be used to indicate N track class clusters obtained by clustering the set of vehicle tracks and the vehicle tracks included in each of the N track class clusters.

Alternatively, the clustering operation may be performed separately for each of the vehicle trajectories in the set of vehicle trajectories according to the initial N sets of clustering center points and the set of points to be classified for each of the vehicle trajectories in the set of vehicle trajectories. The category of each vehicle track in the N track class clusters can be determined according to the Euclidean distance between the N groups of clustering center points and a group of points to be classified of each vehicle track.

In order to improve the accuracy of the clustering result, a condition for ending the clustering is set: after determining the category of each vehicle track in the N track class clusters, updating and determining the clustering center point can be performed, and determining the vehicle track contained in each track class cluster in the N track class clusters repeatedly until the position of the clustering center point is not changed greatly. It should be noted that, the main motivation for clustering the vehicle trajectories through the N groups of clustering center points is to assume that the vehicle motion states in the local subsets of the respective vehicle trajectories are similar in terms of time-space properties.

Through the steps S202 to S206, according to the number N of drivable paths of the target intersection, N reference trajectories are selected from a set of vehicle trajectories of the target intersection, wherein each vehicle trajectory in the set of vehicle trajectories is a section of driving trajectory from the target intersection to the departure of the vehicle corresponding to each vehicle trajectory, and N is a positive integer greater than or equal to 2; respectively selecting a group of clustering center points from each of N reference tracks to obtain initial N groups of clustering center points; according to the initial N groups of clustering center points, clustering operation is carried out on a group of vehicle tracks to obtain a group of clustering results of the vehicle tracks, wherein the clustering results of the group of vehicle tracks are used for indicating N track class clusters obtained by clustering the group of vehicle tracks and the vehicle tracks contained in each track class cluster in the N track class clusters, the problem that the clustering accuracy is poor due to complex intersection environments in the clustering method of the vehicle tracks in the related art is solved, and the clustering accuracy is improved.

In one exemplary embodiment, before selecting N reference trajectories from a set of vehicle trajectories at the target intersection according to the number N of drivable paths at the target intersection, the method further comprises:

s11, acquiring a group of candidate vehicle tracks passing through a target intersection, wherein track information of each candidate vehicle track in the group of candidate vehicle tracks comprises vehicle position information and time information corresponding to each vehicle track;

s12, a group of vehicle tracks is screened out from a group of candidate vehicle tracks according to the characteristics of the target intersection, the vehicle position information and the time information corresponding to each vehicle track.

For a set of vehicle trajectories at a target intersection, it may be determined based on the spatiotemporal information of each vehicle trajectory passing through the target intersection. In this embodiment, before the clustering operation starts, the obtained vehicle track passing through the target intersection may be screened. And screening out the vehicle tracks which do not belong to the target intersection, wherein the screened-out vehicle tracks contain tracks beyond the range of the intersection. Further, considering that the total travel time of the vehicle passing through the target intersection (the time between the start point and the end point of the passing through the target intersection) is subject to the normal distribution, trajectories whose total time passing through the target intersection is greater than and/or less than a certain range can be screened out.

In this embodiment, a set of candidate vehicle trajectories passing through the target intersection may be obtained, and a set of vehicle trajectories may be selected from the set of candidate vehicle trajectories according to characteristics of the target intersection and vehicle position information and time information corresponding to each vehicle trajectory. Here, the track information of each of the candidate vehicle tracks in the set of candidate vehicle tracks may contain vehicle position information and time information corresponding to each of the vehicle tracks. The set of candidate vehicle tracks can be determined by vehicle information detected by a road side unit of the target intersection, positioning data uploaded by the vehicle and the like; the vehicle may directly upload the relevant information to the server and acquire the relevant information from the server.

According to the embodiment, a group of vehicle tracks to be clustered is determined according to the space-time information of the vehicle tracks passing through the target intersection, so that the influence on the selection of the clustering center point caused by the participation of the vehicle tracks which do not belong to the target intersection in the clustering can be avoided, and the accuracy of the clustering result can be improved.

In an exemplary embodiment, selecting a set of cluster center points from each of the N reference trajectories, respectively, to obtain an initial N set of cluster center points, including:

S21, respectively selecting a starting point of each reference track and an ending point of each reference track from each reference track to obtain initial N groups of clustering center points; or,

s22, respectively selecting a starting point of each reference track, an ending point of each reference track and at least one equal-dividing point between the starting point of each reference track and the ending point of each reference track from each reference track to obtain initial N groups of clustering center points.

Considering that the similarity of the vehicle tracks at the intersections is strong, the starting point positions of the vehicles starting from the same lane are similar. In addition, since the vehicles travel to the corresponding lanes, the end positions reached by the vehicles after turning or traveling straight are generally similar, but the turning radii of the vehicles in the middle of the intersection are generally greatly different. In this embodiment, feature points including at least a track start point and a track end point may be used as cluster center points.

In this embodiment, a start point of each reference track and an end point of each reference track may be selected from each reference track, respectively, to obtain an initial N-group cluster center point. The starting point of each reference track, the ending point of each reference track and at least one equal-dividing point between the starting point of each reference track and the ending point of each reference track can be selected from each reference track respectively, so that initial N groups of clustering center points are obtained.

Alternatively, the start point and the end point of each reference trajectory may be determined according to vehicle position information and time information in the vehicle trajectory.

For example, taking initial cluster center points as initial center point a and initial center point B as an example, k tracks may be randomly selected for initializing the center points. Setting both end points (start point, end point) of the k track sample data as an initial center point A and B as different pointsThe initial center point of the track cluster, the number of the center points is 2k, and the initialized center point coordinates are (A _x ,A _y ) And (B) _x ,B _y ) The expression is as shown in formula (1).

Wherein,,and->X, y-axis coordinates representing the start point of the ith vehicle track,/->And->And represents the x, y-axis coordinates of the i-th vehicle track end point.

Alternatively, in the case where the set of cluster center points of each reference trajectory includes a start point, an end point, and at least one bisector point between the start point and the end point of each reference trajectory, the number of bisector points may be determined according to the trajectory characteristics of the vehicle trajectory at the target intersection. In the case where the vehicle trajectory is complex (e.g., there are many drivable paths, there are many actual flow directions of the vehicle, etc.), there may be many equally divided points selected between the start point and the end point as the cluster center points, whereas in the case where the vehicle trajectory is simple, there may be no equally divided points selected or fewer equally divided points selected between the start point and the end point as the cluster center points.

According to the embodiment, the initial position and the end position of the vehicle track of the target intersection are selected as the clustering center point according to the space-time similarity of the initial position and the end position of the vehicle track, the characterization effect of the clustering center point on the vehicle track can be improved, and the accuracy of the clustering result is further improved.

In one exemplary embodiment, a clustering operation is performed on a set of vehicle tracks according to an initial N sets of clustering center points, to obtain a clustering result of the set of vehicle tracks, including:

s31, taking the initial N groups of clustering center points as N groups of current clustering center points, and respectively taking each vehicle track in a group of vehicle tracks as the current vehicle track to execute the following first clustering operation to obtain a clustering result of the group of vehicle tracks:

selecting a group of track points corresponding to N groups of current clustering center points from the current vehicle track to obtain a group of current track points, wherein each current clustering center point in each group of current clustering center points of the N groups of current clustering center points has a one-to-one correspondence with each current track point in the group of current track points;

respectively calculating the sum of the distances between each current clustering center point in each group of current clustering center points and the corresponding current track point in one group of current track points to obtain a distance value corresponding to each group of current clustering center points; determining a track cluster corresponding to a group of current clustering center points with the smallest corresponding distance value from N groups of current clustering center points as a track cluster to which a current vehicle track belongs; or,

Respectively calculating the average value of the distances between each current clustering center point in each group of current clustering center points and the corresponding current track point in one group of current track points to obtain a distance value corresponding to each group of current clustering center points; and determining the track class cluster corresponding to the group of current clustering center points with the smallest corresponding distance value in the N groups of current clustering center points as the track class cluster to which the current vehicle track belongs.

The clustering operation performed on the set of vehicle trajectories may include calculating euclidean distances between the set of clustering center points and a set of trajectory points (i.e., the aforementioned points to be classified) of each vehicle trajectory to perform a division of the trajectory clusters for the set of vehicle trajectories according to the euclidean distances. In this embodiment, when performing the calculation of the correlation of the euclidean distance, the first clustering operation may be performed by using the initial N groups of cluster center points as N groups of current cluster center points, and using each vehicle track in the one group of vehicle tracks as a current vehicle track, to obtain a clustering result of the one group of vehicle tracks.

The first clustering operation may be that a set of track points corresponding to N sets of current cluster center points are selected from the current vehicle tracks to obtain a set of current track points, and the sum of distances between each current cluster center point in each set of current cluster center points and a corresponding current track point in the set of current track points is calculated to obtain a distance value corresponding to each set of current cluster center points, and a track cluster corresponding to a set of current cluster center points with the smallest corresponding distance value in the N sets of current cluster center points is determined as a track cluster to which the current vehicle track belongs.

The first clustering operation may also be that, after obtaining a set of current track points, an average value of distances between each current cluster center point in each set of current cluster center points and a corresponding current track point in a set of current track points is calculated, a distance value corresponding to each set of current cluster center points is obtained, and a track cluster corresponding to a set of current cluster center points with the smallest corresponding distance value in the N sets of current cluster center points is determined as a track cluster to which the current vehicle track belongs. It should be noted that, the distance between the current cluster center point and the corresponding current track point may refer to the euclidean distance.

For example, taking an initial cluster center point as an initial center point A and an initial center point B as an example, a starting point of each sample track to be classified can be obtainedAnd end point->Respectively with two central points A of the kth track cluster _k And B _k The Euclidean distance is calculated and summed (or averaged), and the calculation formula may be shown as formula (2), and the distance value is expressed as distance.

As shown in formula (3), the track cluster k with the smallest distance may be selected as the category of the vehicle track.

k＝arg min(distance(k)) (3)

According to the embodiment, the distance between the clustering center point in each group of clustering center points and the track point of each vehicle track is calculated respectively, the distance between each vehicle track and the reference track is determined, and then the track cluster type of each vehicle track is determined, so that the accuracy of the clustering result can be improved.

In an exemplary embodiment, the clustering operation is performed on a set of vehicle tracks according to the initial N sets of clustering center points, so as to obtain a clustering result of the set of vehicle tracks, and the method further includes:

s41, repeatedly executing the following second clustering operation until the clustering end condition is met, and obtaining a clustering result of a group of vehicle tracks:

taking N groups of clustering center points after the last clustering as N groups of current clustering center points, and respectively taking each vehicle track in a group of vehicle tracks as a current vehicle track to execute a first clustering operation to obtain a clustering result of a group of vehicle tracks, wherein the N groups of clustering center points after the last clustering comprise a group of clustering center points which are respectively determined according to the vehicle tracks contained in each track cluster obtained by the last clustering and correspond to each track cluster;

wherein, the clustering ending condition comprises: the position change of a group of clustering center points of each track cluster in the N track clusters after the current clustering and a group of clustering center points corresponding to the N groups of clustering center points after the last clustering is smaller than or equal to a preset change threshold value.

In order to improve the accuracy of the clustering result, a condition for ending the clustering is set: and determining a new clustering center point according to the result of the first clustering operation, and repeatedly executing the first clustering operation until the clustering center point does not change greatly. In this embodiment, the clustering operation performed on the set of vehicle tracks further includes repeatedly performing the second clustering operation until the clustering end condition is satisfied, to obtain a clustering result of the set of vehicle tracks.

The second clustering operation may include taking N groups of clustered central points after the last clustering as N groups of current clustered central points, and respectively taking each vehicle track in a group of vehicle tracks as a current vehicle track to execute a first clustering operation to obtain a clustering result of the group of vehicle tracks. Here, the N groups of cluster center points after the last clustering may include a group of cluster center points corresponding to each track class cluster, which are determined according to the vehicle track included in each track class cluster obtained by the last clustering.

The clustering end condition may include that a position change between a group of clustering center points of each of the N track class clusters after the current clustering and a corresponding group of clustering center points of the N groups of clustering center points after the previous clustering is less than or equal to a preset change threshold. Here, the preset change threshold may be a preset threshold for determining the change size of the cluster center point.

Alternatively, the position of the cluster center may refer to a coordinate position of the cluster center. Correspondingly, the preset change threshold may be a plurality of change thresholds set according to the number of coordinate axes, may be one change threshold set according to the sum of numerical changes corresponding to different coordinate axes, or may be one change threshold set according to an average value of numerical changes corresponding to different coordinate axes, which is not limited in this embodiment.

According to the embodiment, the clustering center point is updated according to the previous calculation result, and the clustering operation is repeatedly performed until the position change of the clustering center point is smaller than the threshold value, so that the final clustering result is obtained.

In an exemplary embodiment, after performing the first clustering operation with N groups of cluster center points after the last clustering as N groups of current cluster center points and each vehicle track in the group of vehicle tracks as a current vehicle track, the method further includes:

s51, determining a group of clustering center points corresponding to each track class cluster according to the vehicle track contained in each track class cluster obtained after the clustering is finished, and obtaining N groups of clustering center points after the clustering;

s52, determining the position deviation between each group of clustering center points in the N groups of clustering center points after the current clustering and a group of clustering center points corresponding to the N groups of clustering center points after the last clustering, and obtaining the position deviation corresponding to each group of clustering center points;

s53, determining that the clustering ending condition is met under the condition that the sum of the deviation of the position deviations corresponding to the central points of each group of clusters is smaller than or equal to a first deviation threshold value; or,

s54, in a case where the average deviation of the positional deviations corresponding to the cluster center points of each group is less than or equal to the second deviation threshold, it is determined that the cluster end condition is satisfied.

In the process of repeatedly executing the second clustering operation, after the N groups of clustered central points after the last clustering are used as N groups of current clustered central points and each vehicle track in a group of vehicle tracks is used as a current vehicle track to execute the first clustering operation, a group of clustered central points corresponding to each track cluster is determined according to the vehicle track contained in each track cluster obtained after the current clustering is finished, so that N groups of clustered central points after the current clustering are obtained.

Optionally, each group of clustering center points of the N groups of clustering center points after the current clustering is determined after analysis and calculation according to the track points of all the vehicle tracks contained in each track class cluster obtained after the current clustering is finished.

According to the determined positions of N groups of clustering center points after the current clustering, the position deviation of each group of clustering center points in the N groups of clustering center points after the current clustering and a group of clustering center points corresponding to the N groups of clustering center points after the last clustering can be determined, and the position deviation corresponding to each group of clustering center points is obtained.

In the case where the aforementioned preset change threshold is one change threshold (i.e., a first deviation threshold) set according to the sum of numerical value changes corresponding to different coordinate axes, the aforementioned positional deviation corresponding to each group of cluster center points may be compared with the first deviation threshold, and in the case where the deviation sum of positional deviations corresponding to each group of cluster center points is less than or equal to the first deviation threshold, it may be determined that the clustering end condition is satisfied.

In the case where the aforementioned preset variation threshold is one variation threshold (i.e., a second deviation threshold) set according to the numerical variation average value corresponding to the different coordinate axes, the aforementioned positional deviation corresponding to each group of cluster center points may be compared with the second deviation threshold, and in the case where the average deviation of the positional deviations corresponding to each group of cluster center points is less than or equal to the second deviation threshold, it may be determined that the clustering end condition is satisfied.

According to the embodiment, whether the clustering is finished or not is determined according to the relation between the sum of the position deviations of the clustering center points determined by the track type clusters after the clustering and the previous clustering center points or the relation between the average value and the corresponding threshold value, so that the accuracy of the clustering result can be improved.

In an exemplary embodiment, determining a group of cluster center points corresponding to each track class cluster according to the vehicle track contained in each track class cluster obtained after the current clustering, to obtain N groups of cluster center points after the current clustering includes:

s61, determining the center of mass of the starting point of the vehicle track contained in each track class cluster obtained after the current clustering and the center of mass of the ending point of the vehicle track contained in each track class cluster obtained after the current clustering as a group of clustering center points of each track class cluster after the current clustering to obtain N groups of clustering center points after the current clustering; or,

S62, determining the center of mass of the starting point of the vehicle track contained in each track class cluster obtained after the current clustering, the center of mass of the ending point of the vehicle track contained in each track class cluster obtained after the current clustering and the center of mass of the equal division point of the vehicle track contained in each track class cluster obtained after the current clustering as a group of cluster center points of each track class cluster after the current clustering, and obtaining N groups of cluster center points after the current clustering.

When determining N groups of clustering center points after the current clustering according to the vehicle tracks contained in each track class cluster obtained after the current clustering, the centroid of the corresponding track point of the vehicle track contained in each track class cluster obtained after the current clustering can be determined as the corresponding clustering center point.

Correspondingly, when the initial group of clustering center points are the starting point and the ending point of the reference track, the centroid of the starting point of the vehicle track contained in each track class cluster obtained by ending the current clustering and the centroid of the ending point of the vehicle track contained in each track class cluster obtained by ending the current clustering can be determined as a group of clustering center points of each track class cluster after the current clustering, so as to obtain N groups of clustering center points after the current clustering.

For example, taking an initial cluster center point as an initial center point A and an initial center point B as an example, for each category c _k Track cluster l of (2) _i The centroid of the start point and end point of each category may be taken as a new cluster center point as shown in equation (4) and equation (5).

When the initial group of clustering center points are the starting point and the ending point of the reference track and at least one equal dividing point between the starting point and the ending point, the centroid of the starting point of the vehicle track contained in each track class cluster obtained by ending the current clustering, the centroid of the ending point of the vehicle track contained in each track class cluster obtained by ending the current clustering and the centroid of the equal dividing point of the vehicle track contained in each track class cluster obtained by ending the current clustering can be determined as the group of clustering center points of each track class cluster after the current clustering, and N groups of clustering center points after the current clustering are obtained.

According to the embodiment, the clustering center point of the next clustering is determined according to the centroid of the corresponding track point of the vehicle track contained in each track class cluster obtained after each clustering, so that the relevance between the clustering center point and the vehicle track cluster can be improved, and the accuracy of track clustering is further improved.

The clustering method of the vehicle trajectories in the embodiment of the present application is explained below in conjunction with an alternative example. In this alternative example, the target intersection is an intersection, and N is k.

In the alternative example, the vehicle track clustering method for the crossroad is provided, and the clustering center points are set to be a plurality of according to the track characteristics of the vehicle tracks, so that the positions of the vehicle tracks can be expressed more accurately by the clustering center points, and the accuracy of the clustering result is improved.

The clustering method of the vehicle tracks in this alternative example may be as shown in fig. 3, and the flow of the clustering method of the vehicle tracks may include the following steps:

and step 1, determining a clustering classification number k according to traffic scenes and clustering purposes. The traffic scene refers to road environment of an intersection, such as lane layout, driving direction setting and the like; the clustering purpose refers to an index to be finally classified, for example, all travel routes corresponding to respective travel directions classified to intersections, or all travel routes corresponding to respective lanes classified to intersections.

And 2, determining an initial center point of the cluster.

And step 3, classifying the sample track.

And 4, updating the clustering center, and repeating the step 3.

And 5, finishing clustering.

By the optional example, the space-time similarity of the vehicle position and the speed in the tracks from the same starting position to the same ending position is fully utilized, and the vehicle tracks are clustered by selecting a plurality of clustering center points, so that the accuracy of clustering results can be improved, and track features can be extracted more effectively.

It should be noted that, for simplicity of description, the foregoing method embodiments are all expressed as a series of action combinations, but it should be understood by those skilled in the art that the present application is not limited by the order of actions described, as some steps may be performed in other order or simultaneously in accordance with the present application. Further, those skilled in the art will also appreciate that the embodiments described in the specification are all preferred embodiments, and that the acts and modules referred to are not necessarily required in the present application.

From the description of the above embodiments, it will be clear to a person skilled in the art that the method according to the above embodiments may be implemented by means of software plus the necessary general hardware platform, but of course also by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (such as ROM (Read-Only Memory)/RAM (Random Access Memory), magnetic disk, optical disk), including instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method of the embodiments of the present application.

According to another aspect of the embodiments of the present application, there is also provided a vehicle track clustering device for implementing the above vehicle track clustering method. FIG. 4 is a block diagram of an alternative vehicle track clustering apparatus, as shown in FIG. 4, according to an embodiment of the present application, which may include:

a selecting unit 402, configured to select N reference tracks from a set of vehicle tracks of the target intersection according to the number N of drivable paths of the target intersection, where each vehicle track in the set of vehicle tracks is a section of driving track from the target intersection to the departure of the vehicle corresponding to each vehicle track, and N is a positive integer greater than or equal to 2;

a selecting unit 404, connected to the selecting unit 402, configured to select a set of cluster center points from each of the N reference tracks, to obtain an initial N set of cluster center points;

the execution unit 406 is connected to the selection unit 404, and is configured to execute a clustering operation on a set of vehicle tracks according to the initial N sets of clustering center points, to obtain a clustering result of the set of vehicle tracks, where the clustering result of the set of vehicle tracks is used to indicate N track class clusters obtained by clustering the set of vehicle tracks and a vehicle track contained in each track class cluster of the N track class clusters.

It should be noted that, the selecting unit 402 in this embodiment may be used to perform the step S202, the selecting unit 404 in this embodiment may be used to perform the step S204, and the executing unit 406 in this embodiment may be used to perform the step S206.

Through the module, N reference tracks are selected from a group of vehicle tracks of the target intersection according to the number N of the drivable routes of the target intersection, wherein each vehicle track in the group of vehicle tracks is a section of driving track from the target intersection to the departure of the vehicle corresponding to each vehicle track, and N is a positive integer greater than or equal to 2; respectively selecting a group of clustering center points from each of N reference tracks to obtain initial N groups of clustering center points; according to the initial N groups of clustering center points, clustering operation is carried out on a group of vehicle tracks to obtain a group of clustering results of the vehicle tracks, wherein the clustering results of the group of vehicle tracks are used for indicating N track class clusters obtained by clustering the group of vehicle tracks and the vehicle tracks contained in each track class cluster in the N track class clusters, the problem that the clustering accuracy is poor due to complex intersection environments in the clustering method of the vehicle tracks in the related art is solved, and the clustering accuracy is improved.

In an exemplary embodiment, the above apparatus further includes:

an obtaining unit, configured to obtain a set of candidate vehicle tracks passing through the target intersection before selecting N reference tracks from a set of vehicle tracks of the target intersection according to the number N of drivable routes of the target intersection, where track information of each candidate vehicle track in the set of candidate vehicle tracks includes vehicle position information and time information corresponding to each vehicle track;

and the screening unit is used for screening a group of vehicle tracks from a group of candidate vehicle tracks according to the characteristics of the target intersection, the vehicle position information and the time information corresponding to each vehicle track.

In an exemplary embodiment, the selection unit comprises:

the first selection module is used for selecting a starting point of each reference track and an ending point of each reference track from each reference track respectively to obtain initial N groups of clustering center points; or,

the second selecting module is used for selecting a starting point of each reference track, an ending point of each reference track and at least one equal dividing point between the starting point of each reference track and the ending point of each reference track from each reference track respectively to obtain initial N groups of clustering center points.

In one exemplary embodiment, an execution unit includes:

the first execution module is used for taking the initial N groups of clustering center points as N groups of current clustering center points, respectively taking each vehicle track in a group of vehicle tracks as the current vehicle track to execute the following first clustering operation, and obtaining a clustering result of the group of vehicle tracks:

In an exemplary embodiment, the execution unit further comprises:

the second execution module is used for repeatedly executing the following second clustering operation until the clustering end condition is met, so as to obtain a clustering result of a group of vehicle tracks:

In an exemplary embodiment, the above apparatus further includes:

the first determining unit is used for determining a group of clustering center points corresponding to each track type cluster according to the vehicle track contained in each track type cluster obtained after the clustering is finished after taking N groups of clustering center points after the last clustering as N groups of current clustering center points and taking each vehicle track in a group of vehicle tracks as a current vehicle track to execute a first clustering operation, so as to obtain N groups of clustering center points after the clustering;

The second determining unit is used for determining the position deviation between each group of clustering center points in the N groups of clustering center points after the current clustering and a group of clustering center points corresponding to the N groups of clustering center points after the last clustering to obtain the position deviation corresponding to each group of clustering center points;

a third determining unit configured to determine that the clustering end condition is satisfied, in a case where a sum of deviations of the positional deviations corresponding to each group of cluster center points is less than or equal to a first deviation threshold; or,

and a fourth determining unit configured to determine that the clustering end condition is satisfied in a case where an average deviation of the positional deviations corresponding to each group of cluster center points is less than or equal to a second deviation threshold.

In one exemplary embodiment, the first determining unit includes:

the first determining module is used for determining the centroid of the starting point of the vehicle track contained in each track class cluster obtained by ending the current clustering and the centroid of the ending point of the vehicle track contained in each track class cluster obtained by ending the current clustering as a group of clustering center points of each track class cluster after the current clustering to obtain N groups of clustering center points after the current clustering; or,

the second determining module is configured to determine a centroid of a start point of a vehicle track included in each track class cluster obtained by ending the current clustering, a centroid of an end point of a vehicle track included in each track class cluster obtained by ending the current clustering, and a centroid of an equal point of a vehicle track included in each track class cluster obtained by ending the current clustering as a set of cluster center points of each track class cluster after the current clustering, so as to obtain N sets of cluster center points after the current clustering.

It should be noted that the above modules are the same as examples and application scenarios implemented by the corresponding steps, but are not limited to what is disclosed in the above embodiments. It should be noted that the above modules may be implemented in software or in hardware as part of the apparatus shown in fig. 1, where the hardware environment includes a network environment.

According to yet another aspect of embodiments of the present application, there is also provided a storage medium. Alternatively, in the present embodiment, the storage medium may be used to execute a program code of the clustering method of the vehicle track in any one of the above embodiments of the present application.

Alternatively, in this embodiment, the storage medium may be located on at least one network device of the plurality of network devices in the network shown in the above embodiment.

Alternatively, in the present embodiment, the storage medium is configured to store program code for performing the steps of:

s1, selecting N reference tracks from a group of vehicle tracks of a target intersection according to the number N of drivable routes of the target intersection, wherein each vehicle track in the group of vehicle tracks is a section of driving track from the target intersection to the departure of a vehicle corresponding to each vehicle track, and N is a positive integer greater than or equal to 2;

S2, respectively selecting a group of clustering center points from each of N reference tracks to obtain initial N groups of clustering center points;

s3, clustering operation is carried out on a group of vehicle tracks according to the initial N groups of clustering center points, and a clustering result of the group of vehicle tracks is obtained, wherein the clustering result of the group of vehicle tracks is used for indicating N track class clusters obtained by clustering the group of vehicle tracks and vehicle tracks contained in each track class cluster in the N track class clusters.

Alternatively, specific examples in the present embodiment may refer to examples described in the above embodiments, which are not described in detail in the present embodiment.

Alternatively, in the present embodiment, the storage medium may include, but is not limited to: various media capable of storing program codes, such as a U disk, ROM, RAM, a mobile hard disk, a magnetic disk or an optical disk.

According to still another aspect of the embodiments of the present application, there is further provided an electronic device for implementing the clustering method of vehicle tracks described above, where the electronic device may be a server, a terminal, or a combination thereof.

Fig. 5 is a block diagram of an alternative electronic device, according to an embodiment of the present application, including a processor 502, a communication interface 504, a memory 506, and a communication bus 508, as shown in fig. 5, wherein the processor 502, the communication interface 504, and the memory 506 communicate with each other via the communication bus 508, wherein,

A memory 506 for storing a computer program;

the processor 502 is configured to execute the computer program stored in the memory 506, and implement the following steps:

Alternatively, the communication bus may be a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus, or an EISA (Extended Industry Standard Architecture ) bus, or the like. The communication bus may be classified as an address bus, a data bus, a control bus, or the like. For ease of illustration, only one thick line is shown in fig. 5, but not only one bus or one type of bus. The communication interface is used for communication between the electronic device and other equipment.

The memory may include RAM or may include non-volatile memory (non-volatile memory), such as at least one disk memory. Optionally, the memory may also be at least one memory device located remotely from the aforementioned processor.

As an example, the memory 506 may include, but is not limited to, a selection unit 402, a selection unit 404, and an execution unit 406 in the clustering device including the vehicle track. In addition, other module units in the clustering device of the vehicle track may be further included, but are not limited to, and are not described in detail in this example.

The processor may be a general purpose processor and may include, but is not limited to: CPU (Central Processing Unit ), NP (Network Processor, network processor), etc.; but also DSP (Digital Signal Processing, digital signal processor), ASIC (Application Specific Integrated Circuit ), FPGA (Field-Programmable Gate Array, field programmable gate array) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.

Alternatively, specific examples in this embodiment may refer to examples described in the foregoing embodiments, and this embodiment is not described herein.

It will be understood by those skilled in the art that the structure shown in fig. 5 is only schematic, and the device implementing the clustering method of vehicle tracks may be a terminal device, and the terminal device may be a smart phone (such as an Android mobile phone, an iOS mobile phone, etc.), a tablet computer, a palmtop computer, a mobile internet device (Mobile Internet Devices, MID), a PAD, etc. Fig. 5 is not limited to the structure of the electronic device. For example, the electronic device may also include more or fewer components (e.g., network interfaces, display devices, etc.) than shown in FIG. 5, or have a different configuration than shown in FIG. 5.

Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be implemented by a program for instructing a terminal device to execute in association with hardware, the program may be stored in a computer readable storage medium, and the storage medium may include: flash disk, ROM, RAM, magnetic or optical disk, etc.

The foregoing embodiment numbers of the present application are merely for describing, and do not represent advantages or disadvantages of the embodiments.

The integrated units in the above embodiments may be stored in the above-described computer-readable storage medium if implemented in the form of software functional units and sold or used as separate products. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, including several instructions to cause one or more computer devices (which may be personal computers, servers or network devices, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application.

In the foregoing embodiments of the present application, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.

In several embodiments provided in the present application, it should be understood that the disclosed client may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, such as the division of the units, is merely a logical function division, and may be implemented in another manner, for example, multiple units or components may be combined or may be integrated into another system, 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 through some interfaces, units or modules, or may be in electrical or other forms.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution provided in the present embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or at least two units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The foregoing is merely a preferred embodiment of the present application and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present application and are intended to be comprehended within the scope of the present application.

Claims

1. A method of clustering vehicle trajectories, comprising:

selecting N reference tracks from a group of vehicle tracks of a target intersection according to the number N of drivable routes of the target intersection, wherein each vehicle track in the group of vehicle tracks is a section of driving track from the target intersection to the departure of a vehicle corresponding to each vehicle track, and N is a positive integer greater than or equal to 2;

respectively selecting a group of clustering center points from each of the N reference tracks to obtain initial N groups of clustering center points;

And performing clustering operation on the group of vehicle tracks according to the initial N groups of clustering center points to obtain a clustering result of the group of vehicle tracks, wherein the clustering result of the group of vehicle tracks is used for indicating N track class clusters obtained by clustering the group of vehicle tracks and vehicle tracks contained in each track class cluster in the N track class clusters.

2. The method of claim 1, wherein prior to selecting N reference trajectories from a set of vehicle trajectories for a target intersection based on the number N of drivable paths for the target intersection, the method further comprises:

acquiring a set of candidate vehicle tracks passing through the target intersection, wherein track information of each candidate vehicle track in the set of candidate vehicle tracks comprises vehicle position information and time information corresponding to each vehicle track;

and screening the group of vehicle tracks from the group of candidate vehicle tracks according to the characteristics of the target intersection, the vehicle position information and the time information corresponding to each vehicle track.

3. The method of claim 1, wherein the selecting a set of cluster center points from each of the N reference trajectories, respectively, to obtain an initial N set of cluster center points includes:

Respectively selecting a starting point of each reference track and an ending point of each reference track from each reference track to obtain initial N groups of clustering center points; or,

and respectively selecting a starting point of each reference track, an ending point of each reference track and at least one equal dividing point between the starting point of each reference track and the ending point of each reference track from each reference track to obtain the initial N groups of clustering center points.

4. The method of claim 1, wherein the clustering the set of vehicle trajectories according to the initial N sets of cluster centers to obtain a cluster result of the set of vehicle trajectories, comprises:

taking the initial N groups of clustering center points as N groups of current clustering center points, and respectively taking each vehicle track in the group of vehicle tracks as a current vehicle track to execute the following first clustering operation to obtain a clustering result of the group of vehicle tracks:

selecting a group of track points corresponding to the N groups of current clustering center points from the current vehicle track to obtain a group of current track points, wherein each current clustering center point in each group of current clustering center points of the N groups of current clustering center points has a one-to-one correspondence with each current track point in the group of current track points;

Respectively calculating the sum of the distances between each current clustering center point in each group of current clustering center points and the corresponding current track point in the group of current track points to obtain a distance value corresponding to each group of current clustering center points; determining a track cluster corresponding to a group of current clustering center points with the smallest corresponding distance value in the N groups of current clustering center points as the track cluster to which the current vehicle track belongs; or,

respectively calculating the average value of the distances between each current clustering center point in each group of current clustering center points and the corresponding current track point in the group of current track points to obtain a distance value corresponding to each group of current clustering center points; and determining a track cluster corresponding to a group of current clustering center points with the smallest corresponding distance value in the N groups of current clustering center points as the track cluster to which the current vehicle track belongs.

5. The method of claim 4, wherein the clustering operation is performed on the set of vehicle tracks according to the initial N sets of clustering center points to obtain a clustering result of the set of vehicle tracks, and further comprising:

and repeatedly executing the following second clustering operation until the clustering end condition is met, and obtaining a clustering result of the group of vehicle tracks:

Taking the N groups of clustering center points after the last clustering as the N groups of current clustering center points, and respectively taking each vehicle track in the group of vehicle tracks as the current vehicle track to execute the first clustering operation to obtain a clustering result of the group of vehicle tracks, wherein the N groups of clustering center points after the last clustering comprise a group of clustering center points which are respectively determined according to the vehicle tracks contained in each track cluster obtained by the last clustering and correspond to each track cluster;

wherein the clustering end condition includes: the position change of a group of clustering center points of each track cluster in the N track clusters after the current clustering and a group of clustering center points corresponding to the N groups of clustering center points after the last clustering is smaller than or equal to a preset change threshold value.

6. The method of claim 5, wherein after the performing the first clustering operation with the N sets of cluster center points after the last clustering as N sets of current cluster center points and each of the set of vehicle trajectories as a current vehicle trajectory, the method further comprises: determining a group of clustering center points corresponding to each track class cluster according to the vehicle track contained in each track class cluster obtained after the current clustering is finished, and obtaining N groups of clustering center points after the current clustering;

Determining the position deviation between each group of clustering center points in the N groups of clustering center points after the current clustering and a group of clustering center points corresponding to the N groups of clustering center points after the last clustering, and obtaining the position deviation corresponding to each group of clustering center points;

determining that the clustering ending condition is met under the condition that the sum of deviations of the position deviations corresponding to the clustering center points of each group is smaller than or equal to a first deviation threshold value; or,

and determining that the clustering ending condition is met under the condition that the average deviation of the position deviations corresponding to the clustering center points of each group is smaller than or equal to a second deviation threshold value.

7. The method of claim 6, wherein determining a set of cluster center points corresponding to each track class cluster according to the vehicle track included in each track class cluster obtained by ending the current clustering, to obtain the N sets of cluster center points after the current clustering, includes:

determining the center of mass of the starting point of the vehicle track contained in each track class cluster obtained after the current clustering and the center of mass of the ending point of the vehicle track contained in each track class cluster obtained after the current clustering as a group of cluster center points of each track class cluster after the current clustering to obtain N groups of cluster center points after the current clustering; or,

And determining the center of mass of the starting point of the vehicle track contained in each track class cluster obtained after the current clustering, the center of mass of the ending point of the vehicle track contained in each track class cluster obtained after the current clustering and the center of mass of the equal dividing point of the vehicle track contained in each track class cluster obtained after the current clustering as a group of cluster center points of each track class cluster after the current clustering, and obtaining N groups of cluster center points after the current clustering.

8. A clustering device for vehicle trajectories, comprising:

a selection unit, configured to select N reference tracks from a set of vehicle tracks of a target intersection according to a number N of drivable routes of the target intersection, where each vehicle track in the set of vehicle tracks is a section of a driving track from the target intersection to a vehicle leaving the target intersection, where N is a positive integer greater than or equal to 2;

the selecting unit is used for respectively selecting a group of clustering center points from each of the N reference tracks to obtain initial N groups of clustering center points;

the execution unit is used for executing clustering operation on the group of vehicle tracks according to the initial N groups of clustering center points to obtain a clustering result of the group of vehicle tracks, wherein the clustering result of the group of vehicle tracks is used for indicating N track class clusters obtained by clustering the group of vehicle tracks and vehicle tracks contained in each track class cluster in the N track class clusters.

9. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored program, wherein the program when run performs the method of any one of claims 1 to 7.

10. An electronic device comprising a memory and a processor, characterized in that the memory has stored therein a computer program, the processor being arranged to execute the method according to any of claims 1 to 7 by means of the computer program.