Intersection range determination method and device, electronic equipment and medium

Abstract

The invention relates to an intersection range determination method and apparatus, an electronic device and a medium. The method can comprise the steps of obtaining target semantic point cloud data collected in a vehicle driving process; according to the target semantic point cloud data, generating a target road line of a road corresponding to the target semantic point cloud data; other road lines in a first preset range of the target road line are determined, and target intersection points of the target road line and the other road lines are determined; and determining the range of the intersection on the target road line according to the target intersection of the target road line and each other road line. According to the embodiment of the invention, the dependence on data when the intersection range is determined can be reduced; the intersection range can also be determined based on the data representing the road lines, complex data does not need to be used, the application range is wider, and the applicability is higher. In addition, based on simpler data, calculation can be performed more quickly, thereby improving the speed of determining the range of the intersection.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method for determining the range of an intersection, a device for determining the range of an intersection, an electronic device, and a computer-readable storage medium. Background Art

[0002] Road point cloud crowdsourcing data refers to road point cloud data collected through a crowdsourcing platform. Crowdsourcing is a business model that distributes work tasks to the general public who are not specifically targeted, and is widely used in fields such as data collection, software development, and content creation. In the aspect of crowdsourcing collection of road point cloud data, applications on smartphones or other mobile devices are usually used to collect three-dimensional coordinate information of the road surface, and this information can be used in fields such as road detection, map making, and autonomous driving.

[0003] In practical applications, the range of the intersection of a road can be determined based on road point cloud crowdsourcing data. Specifically, according to the stationary situation of the vehicle before the turning segment of the vehicle, it can be identified whether the turning segment occurs at an intersection; if it occurs at an intersection, the starting position of the turning segment is marked as the intersection position, and then the range of the intersection is determined.

[0004] Alternatively, in the preprocessing stage of trajectory data, the original trajectory can be divided into "moving segments" and "staying segments", and the "moving segments" with obvious road geometric features are retained for identifying intersections; the longitude and latitude coordinates of the "moving segments" are subjected to GeoHash encoding to be converted into an encoded trajectory sequence; at the same time, the trajectory activity area is encoded and divided into grids according to the same GeoHash encoding accuracy to obtain a regional encoding matrix; a binary fusion matrix is constructed using the encoded trajectory sequence and the regional encoding matrix, and then a set of feature matrices of intersections and non-intersections is extracted; the KNN (k-Nearest Neighbor) algorithm with a sliding window is used to identify intersections and determine the positions of intersections, and then the range of intersections is determined.

[0005] Or, the intersection stop line can be generated by fitting the points with a speed of 0 in the trajectory data, and the lane boundary lines are cut by the stop line to identify the intersection position, and then a complete intersection surface is generated to complete the intersection identification and further determine the range of the intersection.

[0006] Although the above methods can determine the range of intersections based on road point cloud crowdsourcing data, they need to determine intersections and their ranges based on operations such as parking and turning, and have high requirements for data; when there are no operations such as parking and turning in the data, it is impossible to determine the range of intersections based on road point cloud crowdsourcing data. Summary of the Invention

[0007] In view of the above problems, a method for determining the range of an intersection, a device for determining the range of an intersection, an electronic device, and a computer-readable storage medium are provided to overcome the above problems or at least partially solve the above problems, including:

[0008] A method for determining the range of an intersection, the method including:

[0009] Obtain target semantic point cloud data collected during the driving of a vehicle;

[0010] Generate a target road line of the road corresponding to the target semantic point cloud data according to the target semantic point cloud data;

[0011] Determine other road lines within a first preset range of the target road line, and determine target intersection points of the target road line and each of the other road lines;

[0012] Determine the range of the intersection on the target road line according to the target intersection points of the target road line and each of the other road lines.

[0013] Optionally, the generating the target road line of the road corresponding to the target semantic point cloud data according to the target semantic point cloud data includes:

[0014] Obtain a target trajectory point sequence of the vehicle when collecting the target semantic point cloud data;

[0015] Group the target semantic point cloud data according to the timestamps of the target trajectory point sequence and the timestamps of the target semantic point cloud data to obtain multiple groups of point clouds;

[0016] Generate the target road line according to the first clustering center points of each group of point clouds.

[0017] Optionally, the target trajectory point sequence includes multiple trajectory point data, the target semantic point cloud data includes multiple semantic point cloud data, and the grouping the target semantic point cloud data according to the timestamps of the target trajectory point sequence and the timestamps of the target semantic point cloud data to obtain multiple groups of point clouds includes:

[0018] Use the timestamps of the respective trajectory point data as the second clustering center points, and cluster the multiple semantic point cloud data through the timestamps of the semantic point cloud data to obtain multiple groups of point clouds.

[0019] Optionally, the generating the target road line according to the first clustering center points of each group of point clouds includes:

[0020] Cluster each group of point clouds to determine the first clustering center points of each group of point clouds;

[0021] Perform curve fitting on the first clustering center points of each group of point clouds to generate the target road line.

[0022] Optionally, determining the range of the intersection on the target road line according to the target intersection points of the target road line and each other road line includes:

[0023] Determine the first other intersection points within the second preset range of the target intersection points;

[0024] Generate a first polygon for the intersection on the target road line according to the target intersection points and the first other intersection points;

[0025] Determine the range of the intersection on the target road line according to the first polygon of the intersection on the target road line.

[0026] Optionally, determining the range of the intersection on the target road line according to the target intersection polygon includes:

[0027] Obtain other polygons generated for the intersection corresponding to the first polygon;

[0028] Determine the range of the intersection on the target road line according to the first polygon and the other polygons.

[0029] Optionally, generating a first polygon for the intersection on the target road line according to the target intersection points and the first other intersection points includes:

[0030] Determine the second other intersection points within the third preset range of the target intersection points, where the third preset range is smaller than the second preset range;

[0031] Taking the target intersection points and the second other intersection points as the target clusters, determine the third clustering center points of the target clusters;

[0032] Generate a first polygon for the intersection on the target road line according to the third clustering center points and the first other intersection points; the first other intersection points are the fourth clustering center points.

[0033] Optionally, generating a first polygon for the intersection on the target road line according to the third clustering center points and the first other intersection points includes:

[0034] Connect the third clustering center points and the first other intersection points to generate the first polygon.

[0035] Optionally, determining the range of the intersection on the target road line according to the first polygon of the intersection on the target road line includes:

[0036] Expand the first polygon to obtain a second polygon;

[0037] Determine the range of intersections on the target road line according to the second polygon.

[0038] The present invention also provides a device for determining the range of intersections, and the device includes:

[0039] A point cloud data acquisition module, configured to acquire target semantic point cloud data collected during the driving of the vehicle;

[0040] A road line generation module, configured to generate a target road line of the road corresponding to the target semantic point cloud data according to the target semantic point cloud data;

[0041] An intersection determination module, configured to determine other road lines within a first preset range of the target road line, and determine target intersections between the target road line and each of the other road lines;

[0042] An intersection range determination module, configured to determine the range of intersections on the target road line according to the target intersections between the target road line and each of the other road lines.

[0043] Optionally, the road line generation module is configured to obtain a target trajectory point sequence of the vehicle when collecting the target semantic point cloud data; group the target semantic point cloud data according to the time stamps of the target trajectory point sequence and the time stamps of the target semantic point cloud data to obtain multiple groups of point clouds; generate the target road line according to the first clustering center points of each group of point clouds.

[0044] Optionally, the target trajectory point sequence includes multiple trajectory point data, the target semantic point cloud data includes multiple semantic point cloud data, and the road line generation module is configured to use the time stamps of the respective trajectory point data as the second clustering center points, and cluster the multiple semantic point cloud data through the time stamps of the semantic point cloud data to obtain multiple groups of point clouds.

[0045] Optionally, the road line generation module is configured to cluster each group of point clouds to determine the first clustering center points of each group of point clouds; perform curve fitting on the first clustering center points of each group of point clouds to generate the target road line.

[0046] Optionally, the intersection range determination module is configured to determine first other intersections within a second preset range of the target intersection; generate a first polygon for the intersection on the target road line according to the target intersection and the first other intersections; determine the range of the intersection on the target road line according to the first polygon of the intersection on the target road line.

[0047] Optionally, the intersection range determination module is configured to obtain other polygons generated for the intersections corresponding to the first polygon; and determine the range of the intersections on the target road line according to the first polygon and the other polygons.

[0048] Optionally, the intersection range determination module is configured to determine second other intersections within a third preset range of the target intersection, where the third preset range is smaller than the second preset range; use the target intersection and the second other intersections as a target cluster, and determine a third clustering center point of the target cluster; generate a first polygon for the intersections on the target road line according to the third clustering center point and the first other intersection point, where the first other intersection point is a fourth clustering center point.

[0049] Optionally, the intersection range determination module is configured to connect the third clustering center point and the first other intersection point to generate the first polygon.

[0050] Optionally, the intersection range determination module is configured to expand the first polygon to obtain a second polygon; and determine the range of the intersections on the target road line according to the second polygon.

[0051] The present invention further provides an electronic device, including a processor, a memory, and a computer program stored on the memory and capable of running on the processor. When the computer program is executed by the processor, the method for determining the intersection range as described above is implemented.

[0052] The present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method for determining the intersection range as described above is implemented.

[0053] Advantages of the present invention:

[0054] In the present invention, target semantic point cloud data collected during vehicle driving is obtained; according to the target semantic point cloud data, a target road line of the road corresponding to the target semantic point cloud data is generated; other road lines within a first preset range of the target road line are determined, and target intersection points between the target road line and each of the other road lines are determined; according to the target intersection points between the target road line and each of the other road lines, the range of the intersections on the target road line is determined. Through the embodiments of the present invention, the dependence on data when determining the intersection range can be reduced; the range of the intersections can also be determined based on the data representing the road lines, without using complex data, the application range is wider, and the applicability is higher. In addition, based on simpler data, calculations can be performed faster, thereby improving the speed of determining the range of intersections. Description of the Drawings

[0055] Figure 1 It is a flowchart of steps of a method for determining the range of an intersection according to an embodiment of the present invention;

[0056] Figure 2 It is a flowchart of steps of another method for determining the range of an intersection according to an embodiment of the present invention;

[0057] Figure 3 It is a schematic diagram of a target trajectory point sequence and target semantic point cloud data according to an embodiment of the present invention;

[0058] Figure 4 It is a schematic diagram of a first clustering center point and a target road line according to an embodiment of the present invention;

[0059] Figure 5 It is a schematic diagram of a target intersection point according to an embodiment of the present invention;

[0060] Figure 6 It is a schematic diagram of the range of an intersection according to an embodiment of the present invention;

[0061] Figure 7 It is a flowchart of steps for determining the range of an intersection according to an embodiment of the present invention;

[0062] Figure 8 It is a schematic structural diagram of a device for determining the range of an intersection according to an embodiment of the present invention. Detailed implementation manners

[0063] To make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific implementation manners. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0064] In order to reduce the dependence on data when determining the range of an intersection, the present invention provides a method for determining the range of an intersection. After determining the intersection points of a road line with other road lines, the range of the intersection on the road line can be determined based on these intersection points; compared with the prior art, the present invention can also determine the range of an intersection based on the data representing the road line, without using complex data, has a wider application range, and higher applicability. In addition, based on simpler data, calculations can be performed faster, thereby improving the speed of determining the range of an intersection.

[0065] Referring to Figure 1 , a flowchart of steps of a method for determining the range of an intersection according to an embodiment of the present invention is shown, and it may include the following steps:

[0066] Step 101, obtain target semantic point cloud data collected during the driving of a vehicle.

[0067] Among them, the target semantic point cloud data may refer to the semantic point cloud data collected for the surrounding environment during the driving of the vehicle; the target semantic point cloud data may include position information and semantic information, and the semantic information may include lane lines, road boundary lines, etc.; Exemplarily, the target semantic point cloud data may refer to road point cloud crowdsourcing data.

[0068] In some feasible embodiments, the target semantic point cloud data collected during the driving of the vehicle may be obtained first; Exemplarily, the target semantic point cloud data may include a plurality of semantic point cloud data, and the embodiments of the present invention do not limit this.

[0069] Step 102: Generate a target road line of the road corresponding to the target semantic point cloud data according to the target semantic point cloud data.

[0070] After obtaining the target semantic point cloud data collected during the driving of the vehicle, the target road line of the road corresponding to these target semantic point cloud data can be determined according to these target semantic point cloud data; Exemplarily, the target road line can be used to represent the position of the road corresponding to the target semantic point cloud data. For example: the target road line can be the center line of the road corresponding to the target semantic point cloud data, and the embodiments of the present invention do not limit this.

[0071] Step 103: Determine other road lines within a first preset range of the target road line, and determine the target intersection points of the target road line and each other road line.

[0072] After generating the target road line, search for road data in the vicinity of the target road line through a geometric range; Specifically, other road lines within a first preset range of the target road line can be determined, and the other road lines can be pre-generated road lines.

[0073] After determining the other road lines, the target road line can be geometrically intersected with the other road lines within the first preset range, and the point obtained by this geometric intersection is the target intersection point.

[0074] Step 104: Determine the range of the intersection on the target road line according to the target intersection points of the target road line and each other road line.

[0075] Considering the complexity of the driving route, there will be many different intersection points at the same intersection; Therefore, after determining the target intersection points of the target road line and each other road line, the range of the intersection on the target road line can be determined according to these target intersection points; Exemplarily, an intersection may correspond to multiple target intersection points, and based on these multiple target intersection points, the range of this intersection on the target road line can be determined.

[0076] In an embodiment of the present invention, target semantic point cloud data collected during vehicle driving is obtained; according to the target semantic point cloud data, a target road line of the road corresponding to the target semantic point cloud data is generated; other road lines within a first preset range of the target road line are determined, and target intersection points of the target road line and each of the other road lines are determined; according to the target intersection points of the target road line and each of the other road lines, the range of the intersection on the target road line is determined. Through the embodiment of the present invention, the dependence on data when determining the intersection range can be reduced; the intersection range can also be determined based on the data representing the road line, without using complex data, with a wider application range and higher applicability. In addition, based on simpler data, calculations can be performed faster, thereby improving the speed of determining the intersection range.

[0077] Referring to Figure 2 , a flowchart of steps of another method for determining the intersection range according to an embodiment of the present invention is shown, which may include the following steps:

[0078] Step 201, obtain the target semantic point cloud data collected during vehicle driving.

[0079] In some feasible embodiments, semantic point cloud data may be collected during vehicle driving first and stored; when the range of the intersection needs to be determined, the target semantic point cloud data collected during vehicle driving can be obtained from the stored semantic point cloud data.

[0080] Exemplarily, the obtaining method may be random or sequential according to the time order, and the embodiment of the present invention does not limit this.

[0081] Step 202, obtain the target trajectory point sequence of the vehicle when collecting the target semantic point cloud data.

[0082] In some feasible embodiments, when collecting the target semantic point cloud data, the trajectory points of the vehicle driving may also be collected, and a target trajectory point sequence of the vehicle is generated; the target trajectory point sequence may refer to the sequence generated by the trajectory points on the time axis.

[0083] When determining the range of the intersection, in addition to obtaining the target semantic point cloud data, the target trajectory point sequence of the vehicle when collecting the target semantic point cloud data can also be obtained.

[0084] Exemplarily, as Figure 3 shown, it includes a target trajectory point sequence S-201 and target semantic point cloud data S-202 corresponding to the target trajectory point sequence S-201.

[0085] Step 203: Group the target semantic point cloud data according to the timestamps of the target trajectory point sequence and the target semantic point cloud data to obtain multiple groups of point clouds.

[0086] After obtaining the target trajectory point sequence and the target semantic point cloud data, the timestamps of each trajectory point in the target trajectory point sequence and the timestamps corresponding to each semantic point cloud data in the target semantic point cloud data can be determined.

[0087] After determining the timestamps of each trajectory point in the target trajectory point sequence and the timestamps corresponding to each semantic point cloud data in the target semantic point cloud data, the target semantic point cloud data can be grouped according to the timestamps of each trajectory point in the target trajectory point sequence and the timestamps corresponding to each semantic point cloud data in the target semantic point cloud data, so as to obtain multiple groups of point clouds. At least one semantic point cloud data can be included in one group of point clouds.

[0088] In an embodiment of the present invention, the target trajectory point sequence may include multiple trajectory point data, and the target semantic point cloud data may include multiple semantic point cloud data. When grouping, the following method can be used for grouping:

[0089] Using the timestamps of each trajectory point data as the second clustering center points, cluster the multiple semantic point cloud data through the timestamps of the semantic point cloud data to obtain multiple groups of point clouds.

[0090] In some feasible embodiments, the timestamps of each trajectory point data can be used as the second clustering center points; then, the multiple semantic point cloud data can be clustered through the timestamps of the semantic point cloud data; specifically, the second clustering center point closest to the timestamp of the semantic point cloud data can be found, and the semantic point cloud data can be divided into the group of point clouds corresponding to the second clustering center point.

[0091] As an example, when grouping, the semantics corresponding to the semantic point cloud data can also be combined. For example: group the lane lines in one group and group the road boundary lines in one group. The embodiments of the present invention are not limited thereto.

[0092] Step 204: Generate the target road line according to the first clustering center points of each group of point clouds.

[0093] After obtaining multiple groups of point clouds, the target road line can be generated according to the second clustering center points of the multiple groups of point clouds. Specifically, the target road line can be generated through the following sub-steps, including:

[0094] Sub-step 11: Cluster each group of point clouds to determine the first clustering center points of each group of point clouds.

[0095] First, each group of point cloud sets can be clustered, and the first cluster center point corresponding to each group of point cloud sets can be determined; illustratively, DBSCAN (Density-Based Spatial Clustering of Applications with Noise) can be used to perform clustering and determine the first cluster center point of each group of point cloud sets.

[0096] After obtaining the first cluster center point of each group of point clouds, a sequence of first cluster center points can be obtained, such as Figure 4 As shown, it can be targeted Figure 3 The target semantic point cloud data S-202 in the image is clustered to obtain the first cluster center point S-301.

[0097] Sub-step 12: performing curve fitting on the first cluster center point of each group of point clouds to generate a target road line.

[0098] In some feasible embodiments, after determining the first cluster center point of each group of point cloud sets, curve fitting can be performed on the first cluster center points of the multiple point cloud sets to obtain the target road line; for example, Figure 4 As shown, it can be targeted Figure 4 The first cluster center point S-301 in is subjected to curve fitting to obtain the target road line S-302.

[0099] Step 205: Determine other road lines within a first preset range of the target road line, and determine target intersection points between the target road line and each of the other road lines.

[0100] After the target road line is generated, the road data within the vicinity of the target road line is searched through the geometric range; specifically, other road lines within the first preset range of the target road line may be determined, and the other road lines may be the first generated road lines.

[0101] After determining other road lines, the target road line may be geometrically intersected with other road lines within the first preset range, and the point obtained by the geometric intersection is the target intersection point.

[0102] like Figure 5 As shown, for the target road line S-402, other road lines S-401 within the first preset range of the target road line S-402 can be determined, and the target road line S-402 and each other road line S-401 have a target intersection S-403

[0103] Step 206: Determine a first other intersection within a second preset range of the target intersection.

[0104] In some feasible embodiments, to ensure the accuracy of the results, after determining the target intersection, the first other intersections within the second preset range of the target intersection can be obtained; the second preset range can be set according to the actual situation. For example, it can be 0.1 m. The second preset range can also be an interval range. For example, it can be a range from 0.1 m to 0.3 m from the target intersection. The embodiments of the present invention do not limit this.

[0105] Among them, the first other intersections can refer to the intersections obtained by the semantic point cloud data collected by vehicles in other trips through steps 201 - 205 above. These intersections are within the second preset range of the target intersection and can be referred to as the first other intersections hereafter.

[0106] Step 207: Generate a first polygon for the intersection on the target road line according to the target intersection and the first other intersections.

[0107] In some feasible embodiments, based on this target intersection and the first other intersections, a first polygon can be generated for the intersection on the target road line. The first polygon can be a polygon obtained by connecting the target intersection and the first other intersections, or can be obtained by other means. The embodiments of the present invention do not limit this.

[0108] In an embodiment of the present invention, the first polygon can be generated through the following sub-steps, including:

[0109] Sub-step 21: Determine the second other intersections within the third preset range of the target intersection, where the third preset range is smaller than the second preset range.

[0110] In some feasible embodiments, after determining the target intersection, one or more second other intersections within the third preset range of the target intersection can be determined; the third preset range can be set according to the actual situation. Specifically, the third preset range is smaller than the second preset range. For example, if the second preset range is 0.1 - 0.3 m, the third preset range is less than 0.1 m. The embodiments of the present invention do not limit this.

[0111] Sub-step 22: Use the target intersection and the second other intersections as the target cluster, and determine the third clustering center of the target cluster.

[0112] After determining the target intersection and one or more second other intersections, the target intersection and these one or more second other intersections can be used as a target cluster; then, the third clustering center of the target cluster can be determined.

[0113] Sub-step 23: Generate a first polygon for the intersection on the target road line according to the third clustering center point and the first other intersection point; the first other intersection point is the fourth clustering center point.

[0114] After determining the third clustering center point, a first polygon can be generated for the intersection on the target road line based on the third clustering center point and the first other intersection point; wherein, the first other intersection point can be the fourth clustering center point, that is, based on other semantic point cloud data, through the above-mentioned steps 201 - step 206, and the clustering center points determined from step 21 to sub-step 22.

[0115] As an example, sub-step 23 can be implemented in the following manner:

[0116] Connect the third clustering center point and the first other intersection point to generate the first polygon.

[0117] In some feasible embodiments, the first polygon can be generated by connecting the third clustering center point and the first other intersection point (i.e., the fourth clustering center point). Exemplarily, the third clustering center point or the first other intersection point can be connected in sequence according to the distance, and the embodiments of the present invention are not limited thereto.

[0118] Step 208: Determine the range of the intersection on the target road line according to the first polygon of the intersection on the target road line.

[0119] In some feasible embodiments, after determining the first polygon of the intersection on the target road line, the intersection on the target road line and the range of this intersection can be determined according to this first polygon. After determining this intersection and the range of this intersection, it can be used for subsequent generation of a high-precision map.

[0120] In some other feasible embodiments, after determining the target intersection point, the target road line can be interrupted based on the target intersection point, so that when generating a high-precision map subsequently, it can be recognized that this place is an intersection.

[0121] In an embodiment of the present invention, the range of the intersection on the target road line can be determined through the following sub-steps, including:

[0122] Sub-step 31: Obtain other polygons generated for the intersection corresponding to the first polygon.

[0123] In some feasible embodiments, after generating the first polygon, other polygons generated for the intersection corresponding to the first polygon can be obtained first. These other polygons can be generated based on other point cloud data through the above-mentioned steps 201 - step 207.

[0124] Sub-step 32: Determine the range of intersections on the target road line according to the first polygon and other polygons.

[0125] After obtaining the first polygon and other polygons, the range of intersections on the target road line can be determined according to the first polygon and other polygons; exemplarily, a third polygon can be generated according to the first polygon and other polygons, and then based on this third polygon, the intersections on the target road line and the range of the intersections can be determined.

[0126] In another embodiment of the present invention, determining the range of intersections on the target road line according to the first polygon of the intersections on the target road line includes:

[0127] Sub-step 41: Expand the first polygon to obtain a second polygon.

[0128] In another feasible embodiment, the intersections on the target road line and the range of the intersections can also be determined by expanding the first polygon; exemplarily, the first polygon can be expanded according to a preset rule to obtain a second polygon; the expansion rule can be the expansion length, expansion area, etc., or it can be expanded until it intersects with the lane line point clouds in all directions, and the embodiments of the present invention do not limit this.

[0129] Sub-step 42: Determine the range of intersections on the target road line according to the second polygon.

[0130] After obtaining the second polygon, the intersections on the target road line and the range of the intersections can be determined based on the second polygon; exemplarily, the position corresponding to the second polygon can be used as the intersection, and the range of the second polygon can be used as the range of the intersection.

[0131] As Figure 6 shown, after obtaining the first polygon S-503, the first polygon S-503 can be expanded to obtain a second polygon S-504; the specific expansion rule can be that the first polygon S-503 stops expanding when it intersects with the lane line point clouds S-504 in all directions, and the second polygon S-504 is obtained, and this second polygon S-504 is the range of the intersections on the target road line.

[0132] Exemplarily, as Figure 7 shown, the target semantic point cloud data and the target trajectory point sequence can be obtained first.

[0133] Then, the target semantic point cloud data can be matched to the nearest trajectory points according to the timestamps, so as to obtain multiple groups of point clouds. Specifically, after obtaining the target trajectory point sequence and the target semantic point cloud data, the timestamps of each trajectory point in the target trajectory point sequence and the timestamps corresponding to each semantic point cloud data in the target semantic point cloud data can be determined. After determining the timestamps of each trajectory point in the target trajectory point sequence and the timestamps corresponding to each semantic point cloud data in the target semantic point cloud data, the target semantic point cloud data can be grouped according to the timestamps of each trajectory point in the target trajectory point sequence and the timestamps corresponding to each semantic point cloud data in the target semantic point cloud data, so as to obtain multiple groups of point clouds.

[0134] Next, clustering processing can be sequentially performed on all grouped point clouds to obtain clustering center points.

[0135] After obtaining the clustering centers, curve fitting can be performed on all the clustering center points to obtain the target road line.

[0136] Then, other road lines around the target road line can be extracted; specifically, road data within the vicinity range of the target road line can be found through geometric ranges.

[0137] After obtaining the other road lines, the intersection relationship between the target road line and the other road lines can be calculated. If there is an intersection relationship between the target road line and the other road lines, the intersection points are calculated, the target road line is truncated at the intersection points, and the intersection points are marked as intersection points.

[0138] Next, spatial clustering of the intersections can be performed: points of the same class represent the same intersection. Connect all the points of the same cluster to obtain the first polygon of the intersection. Expand the first polygon until it intersects with the road lane lines in all directions, and the range of the intersection can be obtained. Of course, the first polygon can also be expanded by a certain distance or within a certain range to obtain the range of the intersection.

[0139] Specifically, after determining the target intersection points, one or more second other intersection points within the third preset range of the target intersection points can be determined. After determining the target intersection points and the one or more second other intersection points, the target intersection points and these one or more second other intersection points can be used as a target cluster; then, the third clustering center of the target cluster can be determined. After determining the third clustering center point, a first polygon can be generated for the intersections on the target road line based on the third clustering center point and the first other intersection points.

[0140] After generating the first polygon, other polygons previously generated for the intersection corresponding to the first polygon can be obtained. After obtaining the first polygon and other polygons, the range of the intersection on the target road line can be determined based on the first polygon and other polygons; exemplarily, a third polygon can be generated based on the first polygon and other polygons, and then the intersection on the target road line and the range of the intersection can be determined based on the third polygon.

[0141] In another case, the intersection on the target road line and the range of the intersection can also be determined by expanding the first polygon; exemplarily, the first polygon can be expanded according to a preset rule to obtain a second polygon; the expansion rule can be an expansion length, an expansion area, etc., or can be expanded until it intersects with the lane line point clouds in all directions. After obtaining the second polygon, the intersection on the target road line and the range of the intersection can be determined based on the second polygon; exemplarily, the position corresponding to the second polygon can be used as the intersection, and the range of the second polygon can be used as the range of the intersection.

[0142] In the embodiments of the present invention, the target semantic point cloud data collected during the vehicle driving process can be obtained first; then, when collecting the target semantic point cloud data, the target trajectory point sequence of the vehicle can be obtained; according to the timestamps of the target trajectory point sequence and the target semantic point cloud data, the target semantic point cloud data is grouped to obtain multiple groups of point clouds; the target road line is generated according to the first clustering center points of each group of point clouds; other road lines within the first preset range of the target road line are determined, and the target intersection points of the target road line and each other road line are determined; the first other intersection points within the second preset range of the target intersection points are determined; according to the target intersection points and the first other intersection points, a first polygon is generated for the intersection on the target road line; the range of the intersection on the target road line is determined according to the first polygon of the intersection on the target road line. Through the embodiments of the present invention, the dependence on data when determining the intersection range can be reduced; the intersection range can also be determined based on the data representing the road line, without using complex data, with a wider application range and higher applicability. In addition, based on simpler data, calculations can be performed faster, thereby improving the speed of determining the intersection range.

[0143] It should be noted that for the method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should know that the embodiments of the present invention are not limited by the described action sequence, because according to the embodiments of the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0144] Refer toFigure 8 , showing a schematic structural diagram of a device for determining the range of an intersection according to an embodiment of the present invention, may include the following modules:

[0145] The point cloud data acquisition module 801 is configured to acquire target semantic point cloud data collected during the driving of the vehicle;

[0146] The road line generation module 802 is configured to generate a target road line of the road corresponding to the target semantic point cloud data according to the target semantic point cloud data;

[0147] The intersection determination module 803 is configured to determine other road lines within a first preset range of the target road line, and determine target intersection points between the target road line and each of the other road lines;

[0148] The intersection range determination module 804 is configured to determine the range of the intersection on the target road line according to the target intersection points between the target road line and each of the other road lines.

[0149] In an embodiment of the present invention, the road line generation module 802 is configured to acquire a target trajectory point sequence of the vehicle when collecting the target semantic point cloud data; group the target semantic point cloud data according to the time stamp of the target trajectory point sequence and the time stamp of the target semantic point cloud data to obtain multiple groups of point clouds; generate a target road line according to the first clustering center point of each group of point clouds.

[0150] In an embodiment of the present invention, the target trajectory point sequence includes multiple trajectory point data, and the target semantic point cloud data includes multiple semantic point cloud data. The road line generation module 802 is configured to use the time stamp of each trajectory point data as the second clustering center point, and cluster the multiple semantic point cloud data through the time stamp of the semantic point cloud data to obtain multiple groups of point clouds.

[0151] In an embodiment of the present invention, the road line generation module 802 is configured to cluster each group of point clouds to determine the first clustering center point of each group of point clouds; perform curve fitting on the first clustering center points of each group of point clouds to generate a target road line.

[0152] In an embodiment of the present invention, the intersection range determination module 804 is configured to determine a first other intersection point within a second preset range of the target intersection point; generate a first polygon for the intersection on the target road line according to the target intersection point and the first other intersection point; determine the range of the intersection on the target road line according to the first polygon of the intersection on the target road line.

[0153] In an embodiment of the present invention, the intersection range determination module 804 is configured to acquire other polygons generated for the intersection corresponding to the first polygon; determine the range of the intersection on the target road line according to the first polygon and the other polygons.

[0154] In an embodiment of the present invention, an intersection range determination module 804 is configured to determine a second other intersection within a third preset range of a target intersection, where the third preset range is smaller than the second preset range; use the target intersection and the second other intersection as a target cluster, and determine a third clustering center point of the target cluster; generate a first polygon for the intersections on the target road line according to the third clustering center point and a first other intersection; the first other intersection is a fourth clustering center point.

[0155] In an embodiment of the present invention, the intersection range determination module 804 is configured to connect the third clustering center point and the first other intersection to generate a first polygon.

[0156] In an embodiment of the present invention, the intersection range determination module 804 is configured to expand the first polygon to obtain a second polygon; and determine the range of the intersections on the target road line according to the second polygon.

[0157] In an embodiment of the present invention, target semantic point cloud data collected during the driving of a vehicle is obtained; according to the target semantic point cloud data, a target road line of a road corresponding to the target semantic point cloud data is generated; other road lines within a first preset range of the target road line are determined, and target intersections of the target road line and each of the other road lines are determined; according to the target intersections of the target road line and each of the other road lines, the range of the intersections on the target road line is determined. Through the embodiment of the present invention, the dependence on data when determining the intersection range can be reduced; the range of intersections can also be determined based on data representing road lines, without the need to use complex data, and the application range is wider and the applicability is higher. In addition, based on simpler data, calculations can be performed faster, thereby improving the speed of determining the range of intersections.

[0158] An embodiment of the present invention further provides an electronic device, which may include a processor, a memory, and a computer program stored on the memory and capable of running on the processor. When the computer program is executed by the processor, the method for determining the intersection range as described above is implemented.

[0159] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method for determining the intersection range as described above is implemented.

[0160] For the apparatus embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For related parts, reference may be made to the partial description of the method embodiment.

[0161] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. The same or similar parts among the embodiments may be referred to each other.

[0162] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, an apparatus, or a computer program product. Therefore, the embodiments of the present invention can take the form of an all-hardware embodiment, an all-software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0163] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing terminal devices generate a device for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0164] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0165] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, such that a series of operation steps are executed on the computer or other programmable terminal device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable terminal device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0166] Although the preferred embodiments of the embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the appended claims are intended to be construed as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0167] Finally, it should also be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or terminal device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or terminal device. Without further limitation, an element defined by the phrase "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or terminal device comprising the element.

[0168] The above has introduced in detail a method for determining an intersection range, a device for determining an intersection range, an electronic device and a computer-readable storage medium. Specific examples are used in this text to elaborate on the principles and implementation manners of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A method for determining the intersection range, characterized in that, The method includes: Obtaining target semantic point cloud data collected during vehicle driving; Generating a target road line of the road corresponding to the target semantic point cloud data according to the target semantic point cloud data; Determining other road lines within a first preset range of the target road line, and determining target intersection points between the target road line and each of the other road lines; Determining the range of the intersection on the target road line according to the target intersection points between the target road line and each of the other road lines.

2. The method according to claim 1, wherein The generating a target road line of the road corresponding to the target semantic point cloud data according to the target semantic point cloud data includes: Obtaining a target trajectory point sequence of the vehicle when collecting the target semantic point cloud data; Grouping the target semantic point cloud data according to the timestamps of the target trajectory point sequence and the timestamps of the target semantic point cloud data to obtain multiple groups of point cloud sets; Generating the target road line according to the first clustering center points of each group of point cloud sets.

3. The method according to claim 2, wherein The target trajectory point sequence includes multiple trajectory point data, and the target semantic point cloud data includes multiple semantic point cloud data. The grouping the target semantic point cloud data according to the timestamps of the target trajectory point sequence and the timestamps of the target semantic point cloud data to obtain multiple groups of point cloud sets includes: Using the timestamps of the respective trajectory point data as second clustering center points, and clustering the multiple semantic point cloud data through the timestamps of the semantic point cloud data to obtain multiple groups of point cloud sets.

4. The method according to claim 2, wherein The generating the target road line according to the first clustering center points of each group of point cloud sets includes: Clustering each group of point cloud sets to determine the first clustering center points of each group of point cloud sets; Performing curve fitting on the first clustering center points of each group of point cloud sets to generate the target road line.

5. The method according to claim 1, wherein The determining the range of the intersection on the target road line according to the target intersection points between the target road line and each of the other road lines includes: Determining a first other intersection point within a second preset range of the target intersection point; Generating a first polygon for the intersection on the target road line according to the target intersection point and the first other intersection point; Determining the range of the intersection on the target road line according to the first polygon of the intersection on the target road line.

6. The method according to claim 5, wherein The determining the range of the intersection on the target road line according to the target intersection polygon includes: Obtaining other polygons generated for the intersection corresponding to the first polygon; Determining the range of the intersection on the target road line according to the first polygon and the other polygons.

7. The method according to claim 5, characterized in that The generating a first polygon for the intersection on the target road line according to the target intersection point and the first other intersection point includes: Determining a second other intersection point within a third preset range of the target intersection point, where the third preset range is smaller than the second preset range; Taking the target intersection point and the second other intersection point as a target cluster, and determining a third clustering center point of the target cluster; Generating a first polygon for the intersection on the target road line according to the third clustering center point and the first other intersection point; the first other intersection point is a fourth clustering center point.

8. The method according to claim 7, wherein Generating a first polygon for an intersection on the target road line according to the third clustering center point and the first other intersection point includes: Connecting the third clustering center point and the first other intersection point to generate the first polygon.

9. The method according to claim 5, characterized in that Determining the range of the intersection on the target road line according to the first polygon of the intersection on the target road line includes: Expanding the first polygon to obtain a second polygon; Determining the range of the intersection on the target road line according to the second polygon.

10. A device for determining the intersection range, characterized in that, The apparatus includes: A point cloud data acquisition module, configured to acquire target semantic point cloud data collected during vehicle driving; A road line generation module, configured to generate a target road line of a road corresponding to the target semantic point cloud data according to the target semantic point cloud data; An intersection determination module, configured to determine other road lines within a first preset range of the target road line, and determine target intersection points between the target road line and each of the other road lines; An intersection range determination module, configured to determine the range of the intersection on the target road line according to the target intersection points between the target road line and each of the other road lines.

11. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored on the memory and capable of running on the processor. When the computer program is executed by the processor, it implements the method for determining the intersection range according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, it implements the method for determining the intersection range according to any one of claims 1 to 9.

Images