A lane-level high-precision computable road network construction method, device and electronic equipment

Abstract

The application discloses a lane-level high-precision computable road network construction method and device and electronic equipment, and relates to the computer technology field. The method comprises the following steps: acquiring marking line data and a directional road section reference line, screening a target marking line type, and grouping each marking line in the marking line data; adjusting the vector direction of the marking line to generate endpoint information of the marking line; aggregating the center point and the sub-road section node of the marking line, and sorting the sub-road section node according to the traffic flow direction; after determining the target marking line, performing breaking processing on the target marking line to generate a new marking line and corresponding endpoint information, and constructing a new breakpoint-sub-road section node topology model; according to the marking line endpoint sorting result in the sub-road section node, extracting adjacent marking lines and generating a lane, and aggregating the lane into a lane group based on the order of the sub-road section node; and according to the topological relationship between the lanes and the lane groups, supplementing lane connectors between the lanes. The application has high efficiency and high precision, and can be widely applied to the computer technology field.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, and in particular to a lane-level high-precision computable road network construction method and device and electronic equipment. BACKGROUND

[0002] In recent years, with the rise of intelligent transportation, various simulation, path planning, and traffic analysis algorithms have gradually improved, and the application scenarios of intelligent transportation systems have also expanded. The computable road network is the cornerstone of the effective operation of the intelligent transportation system, and its mapping and editing work is a time-consuming and labor-intensive task in system building, directly affecting the operation effect of the system. Therefore, higher demands are placed on the process, quality, update cycle, and information completeness of the computable road network information. More intuitively, the computable road network information needs to balance the matching degree with the actual road conditions and the complete topology required for traffic analysis.

[0003] Currently, through point cloud processing, remote sensing image processing, and various production platforms, the industry has achieved high-precision collection of road surface information, such as accurate calibration of road marking basic information, including color, line type, and marking type information.

[0004] The existing technology mainly focuses on high-precision data production of road visualization information (such as road markings, ground arrows, etc.), but there is less extraction and description of the correlation between markings and the traffic flow characteristics they represent. It is unable to accurately extract lane-level computable road networks, correctly reflect road traffic conditions and traffic flow organization characteristics, and thus cannot well serve the simulation, analysis, and other functional scenarios of intelligent transportation systems.

[0005] At the same time, the existing technology generally achieves editing of the spatial geometry and rule attributes of the computable road network through manual processing, which is labor-intensive and inefficient. SUMMARY

[0006] Therefore, the present application provides a lane-level high-precision computable road network construction method, device, and electronic equipment that is efficient and high-precision.

[0007] An aspect of an embodiment of the present application provides a lane-level high-precision computable road network construction method, comprising:

[0008] Obtaining marking data and directed road segment reference lines, screening target marking types, and grouping each marking in the marking data;

[0009] Adjusting the vector direction of the marking to generate endpoint information of the marking; wherein the endpoint information includes start point information and end point information;

[0010] Performing first clustering processing on each endpoint of the marking to obtain a center point of the marking;

[0011] performing a second clustering process on the center points to obtain sub-road segment nodes, and sorting the sub-road segment nodes according to traffic flow directions;

[0012] After determining the target marking line according to the target marking line type, performing a breaking process on the target marking line according to the sub-road segment nodes, generating a new marking line and corresponding end point information, and constructing a new topology model of breakpoint-sub-road segment nodes;

[0013] performing clustering and sorting on the marking line end points in a single sub-road segment node;

[0014] According to the sorting results of the marking line end points in the sub-road segment node, extracting adjacent marking lines and generating lanes, and aggregating the lanes into a lane group based on the order of the sub-road segment nodes;

[0015] According to the topological relationship between the lanes and the lane groups, supplementing lane connectors between the lanes.

[0016] Optionally, in the step of obtaining marking line data and directional road segment reference lines, screening target marking line types, and grouping each marking line in the marking line data,

[0017] The data structure of the marking line data is:

[0018] L = {geom, type, R}

[0019] wherein L represents marking line data; geom is vector data; type is marking line type information; and R is an associated directional road segment reference line;

[0020] The target marking line types include a crossable opposite lane boundary line, a crossable same direction lane boundary line, a tidal lane line, a lane edge line, a harbor-style stop marking line, a prohibited crossable opposite lane boundary line, a prohibited crossable same direction lane boundary line, a bus lane line, and a variable guide lane marking line;

[0021] The data structure of the directional road segment reference line is:

[0022] R = {geom}

[0023] wherein R is an associated directional road segment reference line; and geom is vector data.

[0024] Optionally, the step of adjusting the vector direction of the marking line and generating end point information of the marking line includes:

[0025] According to the grouping of each marking line in the marking line data, processing each directional road segment and the associated marking line, adjusting the vector direction of the marking line according to the traffic flow direction, generating marking line end point information at the geometric start and end points, and constructing a corresponding topological association model.

[0026] The data structure of the end point of the marking line is:

[0027] P = {lon, lat, se, L}

[0028] P is the end point of the marking line; lon and lat are the longitude and latitude of the end point; se is the start-end point attribute identifier of the end point; and L is the end point associated marking line information.

[0029] Optionally, the first clustering processing is performed on each end point of the marking line, and a center point of the marking line is aggregated, including:

[0030] The end points of the marking line are clustered, and the center point is aggregated;

[0031] According to the end point information of the marking line, the DBSCAN density clustering algorithm is used to cluster the end points of the marking line, and a clustering cluster is obtained; wherein the neighborhood between each end point is obtained through the Euler distance between two points;

[0032] For each clustering cluster, the centroid of the clustering cluster is selected as the center point for recording.

[0033] Optionally, the second clustering processing is performed on the center point to obtain a sub-road segment node, and the sub-road segment node is sorted according to the traffic flow direction, including:

[0034] According to the directional road reference line and the center point of the marking line, a projection operation is performed on the center point;

[0035] The center point is secondarily clustered according to the on-line distance of the projection point, and a sub-road segment node set is obtained;

[0036] According to the on-line distance between the projection point on the directional road reference line and the vector start point, the center point is sequentially numbered from small to large, wherein the sequence number starts from 1.

[0037] Optionally, the marking line end points in a single sub-road segment node are clustered and sorted, including:

[0038] According to the newly constructed breakpoint-sub-road segment node topological model, the DBSCAN density clustering algorithm is used to cluster the end points in the sub-road segment node; wherein the neighborhood of each end point is determined through the Euler distance between two points;

[0039] According to the distance between the clustering cluster centroid and the directional road reference line, the order of the clustering cluster is determined from small to large; wherein the distance on the left side of the line segment is negative, and the distance on the right side of the line segment is positive.

[0040] Optionally, according to the sorting result of the marking line end points in the sub-road segment node, adjacent marking lines are extracted and lanes are generated, and the lanes are aggregated into a lane group based on the order of the sub-road segment node, including:

[0041] According to the end point ordering result in the sub-section node, the target marking line in the adjacent two sub-section nodes is ordered according to the end point order of the start and end points;

[0042] Lane lines between adjacent marking lines are generated according to the order;

[0043] According to the order relationship of the sub-section nodes, the lanes are aggregated into a series of lane groups in an upstream and downstream continuous manner;

[0044] According to the order, lane lines between adjacent marking lines are generated, including:

[0045] The shorter marking line of the left and right marking lines is selected as a main marking line, when the left and right marking lines are of the same length, the right marking line is selected as the main marking line, the other marking line outside the main marking line is selected as a reference line, the perpendicular lines of each control point of the main marking line to the reference line are sequentially drawn, and the midpoints of the perpendicular line segments are selected as control points of the lane line, the target lane line is sequentially generated by indentation according to a predetermined rule, and the lane type is classified and the target section attribute information is corrected according to the marking line type information in the marking line data.

[0046] Another aspect of the embodiment of the application further provides a lane-level high-precision computable road network construction device, comprising:

[0047] A first module is configured to acquire marking line data and a directional section reference line, filter target marking line types, and group each marking line in the marking line data;

[0048] A second module is configured to adjust the vector direction of the marking line and generate end point information of the marking line; wherein the end point information comprises start point information and end point information;

[0049] A third module is configured to perform first clustering processing on each end point of the marking line, and aggregate to obtain a center point of the marking line;

[0050] A fourth module is configured to perform second clustering processing on the center point to obtain a sub-section node, and sort the sub-section node according to the traffic flow direction;

[0051] A fifth module is configured to determine a target marking line according to the target marking line type, perform breaking processing on the target marking line according to the sub-section node, generate new marking lines and corresponding end point information, and construct a new breakpoint-sub-section node topology model;

[0052] A sixth module is configured to sort the marking line end points in a single sub-section node after clustering;

[0053] a seventh module configured to extract adjacent lane marks and generate lanes according to the sorting result of the lane mark endpoints in the sub-road segment node, aggregate the lanes into a lane group based on the order of the sub-road segment node;

[0054] an eighth module configured to supplement lane connectors between lanes according to the topological relationship between the lanes and the lane groups.

[0055] Another aspect of the embodiment of the present application further provides an electronic device, including a processor and a memory.

[0056] The memory is configured to store a program.

[0057] The processor executes the program to implement the method as described above.

[0058] Another aspect of the embodiment of the present application further provides a computer readable storage medium, which stores a program, and the program is executed by a processor to implement the method as described above.

[0059] The embodiment of the present application further discloses a computer program product or a computer program, which comprises computer instructions stored in a computer readable storage medium. A processor of a computer device can read the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to enable the computer device to execute the method as described above.

[0060] The embodiment of the present application acquires lane mark data and directed road segment reference lines, filters target lane mark types, and groups each lane mark in the lane mark data; adjusts the vector direction of the lane mark to generate endpoint information of the lane mark; wherein the endpoint information comprises start point information and end point information; performs first clustering processing on each endpoint of the lane mark to aggregate a center point of the lane mark; performs second clustering processing on the center point to obtain a sub-road segment node, and sorts the sub-road segment node according to a traffic flow direction; after determining a target lane mark according to the target lane mark type, performs breaking processing on the target lane mark according to the sub-road segment node to generate a new lane mark and corresponding endpoint information, and constructs a new topological model of the breakpoint-sub-road segment node; sorts the lane mark endpoints in a single sub-road segment node after clustering; extracts adjacent lane marks and generates lanes according to the sorting result of the lane mark endpoints in the sub-road segment node, aggregates the lanes into a lane group based on the order of the sub-road segment node; and supplements lane connectors between lanes according to the topological relationship between the lanes and the lane groups. The present application has high efficiency and high precision. BRIEF DESCRIPTION OF DRAWINGS

[0061] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort.

[0062] Figure 1 is the overall flowchart of the automatic editing of the reticle based on the computable road network of the present application;

[0063] Figure 2 is the original data schematic diagram of the specific implementation case of the present application;

[0064] Figure 3 is the density clustering neighborhood schematic diagram of the present application;

[0065] Figure 4 is the reconstructed reticle-endpoint data schematic diagram of the specific implementation case of the present application;

[0066] Figure 5 is the schematic diagram of generating a lane of the present application;

[0067] Figure 6 is the schematic diagram of generating a lane connector of the present application;

[0068] Figure 7 is the schematic diagram of the final construction result of the specific implementation case of the present application. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solutions and advantages of the present application more clear, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0070] In view of the problems existing in the prior art, one aspect of the embodiment of the present application provides a lane-level high-precision computable road network construction method, comprising:

[0071] Obtaining reticle data and a directed road section reference line, screening a target reticle type, and grouping each reticle in the reticle data;

[0072] Adjusting the vector direction of the reticle to generate endpoint information of the reticle; wherein the endpoint information includes start point information and end point information;

[0073] Performing first clustering processing on each endpoint of the reticle to obtain a center point of the reticle;

[0074] Performing second clustering processing on the center point to obtain a sub-road section node, and sorting the sub-road section node according to the traffic flow direction;

[0075] After determining the target marking line according to the target marking line type, the target marking line is broken according to the sub-road segment node, a new marking line and corresponding endpoint information are generated, and a new breakpoint-sub-road segment node topology model is constructed;

[0076] The marking line endpoints in a single sub-road segment node are clustered and sorted;

[0077] According to the sorting result of the marking line endpoints in the sub-road segment node, adjacent marking lines are extracted and lanes are generated, and the lanes are aggregated into a lane group based on the order of the sub-road segment node;

[0078] According to the topological relationship between the lanes and the lane groups, lane connectors between the lanes are supplemented.

[0079] Optionally, in the step of acquiring marking line data and a directed road segment reference line, screening a target marking line type, and grouping each marking line in the marking line data,

[0080] The data structure of the marking line data is:

[0081] L={geom,type,R}

[0082] Wherein, L represents marking line data; geom is vector data; type is marking line type information; and R is an associated directed road segment reference line;

[0083] The target marking line type includes a crossable opposite lane boundary line, a crossable same lane boundary line, a tidal lane line, a lane edge line, a harbor-type stop marking line, a prohibited crossable opposite lane boundary line, a prohibited crossable same lane boundary line, a bus lane line, and a variable guide lane marking line;

[0084] The data structure of the directed road segment reference line is:

[0085] R={geom}

[0086] Wherein, R is an associated directed road segment reference line; and geom is vector data.

[0087] Optionally, the step of adjusting the vector direction of the marking line and generating endpoint information of the marking line includes:

[0088] According to the grouping condition of each marking line in the marking line data, each directed road segment and the associated marking line are processed, the vector direction of the marking line is adjusted according to the traffic flow direction, the marking line endpoint information is generated at the geometric start and end points, and a corresponding topological association model is constructed;

[0089] Wherein, the data structure of the marking line endpoint is:

[0090] P={lon,lat,se,L}

[0091] Wherein, P is the end point of the marking line; lon, lat are the longitude and latitude of the end point; se is the start and end point attribute identification of the end point; L is the end point associated marking line information.

[0092] Optionally, the first clustering processing is performed on each end point of the marking line, and a center point of the marking line is aggregated, comprising:

[0093] The end points of the marking line are clustered, and a center point is aggregated;

[0094] According to the end point information of the marking line, DBSCAN density clustering algorithm is used to cluster the end points of the marking line, and a clustering cluster is obtained; wherein the neighborhood between each end point is obtained through the Euler distance between two points;

[0095] For each clustering cluster, the centroid of the clustering cluster is selected as the center point for recording.

[0096] Optionally, the second clustering processing is performed on the center point to obtain a sub-road segment node, and the sub-road segment node is sorted according to the traffic flow direction, comprising:

[0097] According to the directed road reference line and the center point of the marking line, a projection operation is performed on the center point;

[0098] The center point is secondarily clustered according to the on-line distance of the projection point, and a sub-road segment node set is obtained;

[0099] According to the on-line distance between the projection point on the directed road reference line and the vector start point, the center point is sequentially numbered from small to large, wherein the sequence number starts from 1.

[0100] Optionally, the marking line end points in a single sub-road segment node are clustered and sorted, comprising:

[0101] According to the newly constructed breakpoint-sub-road segment node topology model, DBSCAN density clustering algorithm is used to cluster the end points in the sub-road segment node; wherein the neighborhood of each end point is determined through the Euler distance between two points;

[0102] According to the distance between the clustering cluster centroid and the directed road reference line, the order of the clustering cluster is determined from small to large; wherein the left side distance is negative and the right side distance is positive.

[0103] Optionally, according to the sorting result of the marking line end points in the sub-road segment node, adjacent marking lines are extracted and lanes are generated, and the lanes are aggregated into a lane group based on the order of the sub-road segment node, comprising:

[0104] According to the sorting result of the marking line end points in the sub-road segment node, the target marking lines in adjacent two sub-road segment nodes are sorted according to the end point order of the start and end points;

[0105] generating lane lines between adjacent markings according to the order;

[0106] aggregating lanes into a series of lane groups in an upstream and downstream continuous manner according to the order relationship of the sub-road segment nodes;

[0107] The method comprises the following steps:

[0108] The shorter marking of the left and right markings is selected as the main marking, the right marking is selected as the main marking when the left and right markings are of the same length, the other marking other than the main marking is selected as the reference line, the perpendicular lines are drawn from each control point of the main marking to the reference line in sequence, the midpoints of the perpendicular line segments are selected as the control points of the lane lines, the target lane lines are generated by sequentially connecting the control points according to a predetermined rule, and the lane types are classified and the target road segment attribute information is corrected according to the marking type information in the marking data.

[0109] Another aspect of the embodiment of the present application also provides a lane-level high-precision computable road network construction device, which comprises:

[0110] The first module is configured to acquire marking data and a directional road segment reference line, filter target marking types, and group each marking in the marking data;

[0111] The second module is configured to adjust the vector direction of the marking and generate endpoint information of the marking; wherein the endpoint information comprises start point information and end point information;

[0112] The third module is configured to perform first clustering processing on each endpoint of the marking to obtain a center point of the marking;

[0113] The fourth module is configured to perform second clustering processing on the center point to obtain a sub-road segment node, and sort the sub-road segment node according to the traffic flow direction;

[0114] The fifth module is configured to determine a target marking according to the target marking type, perform breaking processing on the target marking according to the sub-road segment node to generate a new marking and corresponding endpoint information, and construct a new breakpoint-sub-road segment node topology model;

[0115] The sixth module is configured to sort the marking endpoints in a single sub-road segment node after clustering the marking endpoints;

[0116] The seventh module is configured to extract adjacent markings and generate lanes according to the sorting result of the marking endpoints in the sub-road segment node, and aggregate the lanes into lane groups based on the order of the sub-road segment nodes;

[0117] The eighth module is configured to supplement lane connectors between lanes according to the topological relationship between the lanes and the lane groups.

[0118] Another aspect of the embodiment of the present application further provides an electronic device, comprising a processor and a memory;

[0119] The memory is used for storing a program;

[0120] The processor executes the program to realize the method as described above.

[0121] Another aspect of the embodiment of the present application further provides a computer readable storage medium, which stores a program, and the program is executed by a processor to realize the method as described above.

[0122] The embodiment of the present application further discloses a computer program product or a computer program, which comprises computer instructions stored in a computer readable storage medium. A processor of a computer device can read the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method as described above.

[0123] The specific implementation process of the present application will be described in detail below with reference to the accompanying drawings of the specification:

[0124] The present application is a lane-level computable road network automatic editing method based on high-precision marking data, which can be realized by the following steps:

[0125] (1) reading high-precision marking data and directional road segment reference lines, screening target marking types, and grouping the marking lines according to the associated directional road segments;

[0126] (2) adjusting the vector direction of the marking line, and generating the starting and ending point information of the marking line;

[0127] (3) clustering the marking line endpoints and aggregating the center points;

[0128] (4) performing secondary clustering on the center points to obtain sub-road segment nodes, and sorting them according to the traffic flow direction;

[0129] (5) based on the sub-road segment nodes, the target marking line is broken, new marking lines and endpoints are formed, and a topology model is reconstructed;

[0130] (6) clustering the marking line endpoints in a single sub-road segment node and sorting them;

[0131] (7) based on the endpoint sorting in the sub-road segment node, adjacent marking lines are extracted and lanes are generated, and the lanes are aggregated into lane groups according to the order of the sub-road segment nodes;

[0132] (8) according to the topological relationship between the lanes and the lane groups, lane connectors between the lanes are supplemented.

[0133] The step (1) is specifically reading high-precision marking line data and directional road section reference line, screening target marking line type, and grouping marking lines according to associated directional road sections. The original marking line data is defined as structure L:

[0134] L={geom, type, R}

[0135] Wherein, geom is marking line vector data, type is marking line type information, and R is associated directional road section reference line information. The target marking line types involved in the application include crossable opposite lane boundary line, crossable same direction lane boundary line, tidal lane line, lane edge line, harbor type bus stop marking line, non-crossable opposite lane boundary line, non-crossable same direction lane boundary line, bus lane line, and variable guide lane marking line. The directional road section reference line is vector line data which can basically reflect traffic flow direction and road geometry. The marking line type and associated directional road section information are all contained in the high-precision marking line data.

[0136] The original directional road section reference line data is defined as structure R:

[0137] R={geom}

[0138] Wherein, geom is vector data of the directional road section reference line.

[0139] The step (2) is specifically adjusting marking line vector direction and generating marking line start and end point information. According to the grouping in (1), each directional road section and its associated marking line is processed, the marking line vector direction is adjusted according to traffic flow direction, marking line end point information is generated at the geometric start and end points, and a corresponding topological association model is constructed. The end point data is defined as structure P:

[0140] P={lon, lat, se, L}

[0141] Wherein, lon and lat are longitude and latitude of the end point, se is start and end point attribute identification of the end point, and L is end point associated marking line information.

[0142] The step (3) is specifically clustering marking line end points and aggregating center points. According to the marking line end points {P i} generated in (2), DBSCAN density clustering algorithm is used to cluster the marking line end points, and clustering clusters {C p} are obtained, wherein the eps neighborhood of each end point is obtained through the Euler distance d e (a, b) of two points a(x a , y a ) and b(x b , y b ). e(a, b), the straight-line distance between two points, is calculated as formula (1), the neighborhood radius is e, and the minimum number of points in the neighborhood of the core object is MinPts. For each cluster C p , the center point record is selected as the centroid, and the centroid is calculated as formula (2).

[0143]

[0144]

[0145]

[0146] (x i ,y i )∈C (2)

[0147] The step (4) is specifically: performing secondary clustering on the center points to obtain sub-road segment nodes, and sorting according to the traffic flow direction. According to the center points generated in (2) and the directional road segment reference line read in (1), the projection operation is performed on the center points, and the center points are sorted according to the on-line distance d l (p, q) of the projection point p on the line. i The secondary clustering is performed on the center points to obtain the sub-road segment node set {BP s}, and at the same time, the center points are sequentially numbered from 1 according to the on-line distance d l (p, v s ) of the projection point p on the line and the vector starting point v i .

[0148] The step (5) is specifically: based on the sub-road segment nodes, the target marking line is broken, new marking lines and end points are formed, and the topological model is reconstructed. According to the sub-road segment nodes {BP e} generated in (4) and the target marking line-end point model generated in (2), the start and end points of the marking line and the order of the associated sub-road segment nodes are analyzed, the target marking line whose start and end points are not adjacent is broken, and new marking line-end point data is formed. The breakpoint position is selected as the foot of the sub-road segment node on the target marking line.

[0149] The step (6) is specifically: clustering and sorting the marking line end points in a single sub-road segment node. According to the breakpoint-sub-road segment node relationship model newly generated in (5), the DBSCAN density clustering algorithm is used to cluster the end points in the sub-road segment node, wherein the eps neighborhood of each end point is obtained through the Euler distance d node (a, b) between two points, the neighborhood radius is e nodeMeanwhile, the order of the clustering cluster is determined from small to large according to the distance between the clustering cluster center and the reference line of the directed road section, wherein the distance on the left side of the line segment is negative and the distance on the right side of the line segment is positive.

[0150] The step (7) is specifically: based on the end point order in the sub-road section node, adjacent marking lines are extracted and lanes are generated, and the lanes are aggregated into a lane group based on the order of the sub-road section node. According to the end point order of the end points in the sub-road section node generated in (6), the target marking lines in the adjacent two sub-road section nodes are sorted according to the end point order of the start and end points, and the lane lines between the adjacent marking lines are generated in sequence according to the sorting, and the generation mode is that the shorter marking line of the left and right marking lines is selected as the main marking line L major , the right marking line is selected preferentially when the lengths are the same, the other marking line is the reference line L refer , the perpendicular lines of each control point of L major are sequentially made to the reference line L refer , and the midpoints of the perpendicular line segments are selected as the control points of the lane lines, the control points are indented according to a predetermined rule, and the target lane lines are sequentially connected and generated, and the lane types are classified and the target road section attribute information is corrected according to the marking line type information in the marking line data. Meanwhile, the lanes are aggregated into a series of lane groups in an upstream and downstream continuous mode according to the order relationship of the sub-road section nodes. The lane data is defined as structure Lane:

[0151] Lane={geom,type,Lline,Rline,Link,laneindex}

[0152] Wherein, geom is the lane line vector data, type is the lane line type information, Lline and Rline are left and right associated marking line information respectively, Link is the lane group information, and laneindex is the lane order.

[0153] The lane group data is defined as structure Link:

[0154] Link={fbp,tbp,linkindex,R}

[0155] Wherein, fbp and tbp are associated start and end sub-road section node information respectively, linkindex is the lane group order, and R is the associated directed road section reference line.

[0156] The step (8) is specifically: according to the topological relationship between the lanes and the lane groups, lane connectors between the lanes are supplemented. According to the lane geometry and the lane group topological relationship generated in (7), the traffic flow organization relationship between the lanes is judged, and the corresponding lane connector is generated, so that the calculable road network is kept connected in geometry and logic, and the geometry of the lane connector is generated by a quadratic Bezier curve and a cubic Bezier curve. The calculation method of the quadratic Bezier curve is shown in formula (3), and the calculation method of the cubic Bezier curve is shown in formula (4).

[0157] LC = (1 - t) 2 P0 + 2t(1 - t)P0' + t 2 P1, t e [0, 1] (3)

[0158] LC = (1 - t) 3 P0 + 3t(1 - t) 2 P0' + 3t 2 (1 - t)P1' + t3P1, t e [0, 1] (4)

[0159] Wherein, P0, P1 are end points of upstream and downstream lanes, P'0, P'1 are selected control points.

[0160] The specific implementation process of the present application will be described in detail below with the implementation process of a specific application scenario as an example:

[0161] As shown in the flowchart of a specific embodiment of the present application based on automatic editing of a mark line calculable road network. Referring to Figure 1 , the specific steps of the present embodiment include: Figure 1

[0162] (1) Read high-precision mark line data and directed road segment reference lines, filter target mark line types, and group mark lines according to associated directed road segments;

[0163] (2) Adjust the vector direction of the mark line, and generate mark line start and end point information;

[0164] (3) Cluster the mark line end points, and aggregate the center points;

[0165] (4) Secondary cluster the center points to obtain sub-road segment nodes, and sort them according to traffic flow direction;

[0166] (5) Based on the sub-road segment nodes, break the target mark line, form new mark lines and end points, and reconstruct the topology model;

[0167] (6) Cluster the mark line end points within a single sub-road segment node, and sort them;

[0168] (7) Based on the end point sorting within the sub-road segment node, extract adjacent mark lines and generate lanes, and aggregate the lanes into lane groups based on the order of the sub-road segment nodes;

[0169] (8) According to the topological relationship between the lanes and lane groups, supplement the lane connectors between the lanes.

[0170] Step (1) is: reading high-precision mark line data and directed road segment reference lines, filtering target mark line types, and grouping mark lines according to associated directed road segments. The original mark line data is defined as structure L:​

[0171] L = {geom, type, R}

[0172] Wherein, geom is the vector data of marking line, type is the type information of marking line, R is the reference line information of associated directional road section, the target marking line types involved in the application include the marking line which can cross the boundary line of opposite carriageway, the marking line which can cross the boundary line of same direction carriageway, the marking line of tidal carriageway, the marking line of carriageway edge, the marking line of harbor type bus stop, the marking line which prohibits crossing the boundary line of opposite carriageway, the marking line which prohibits crossing the boundary line of same direction carriageway, the marking line of bus lane, and the marking line of variable direction carriageway. The directional road section reference line is the vector line data which can basically reflect the traffic flow direction and the road geometry. The marking line type and the associated directional road section information are all included in the information of high-precision marking line data.

[0173] The original directional road section reference line data is defined as structure R:

[0174] R = {geom}

[0175] Wherein, geom is the vector data of directional road section reference line.

[0176] Step (2) is to adjust the vector direction of marking line and generate the starting and ending point information of marking line. According to the grouping in (1), each directional road section and its associated marking line is processed respectively, the vector direction of marking line is adjusted according to the traffic flow direction, the end point information of marking line is generated at the geometric starting and ending point, and the corresponding topological association model is constructed. The end point data is defined as structure P:

[0177] P = {lon, lat, se, L}

[0178] Wherein, lon and lat are the longitude and latitude of end point, se is the starting and ending point attribute identification of end point, and L is the end point associated marking line information.

[0179] As shown in Figure 2 , it is a schematic diagram of high-precision marking line and directional road section reference line at a certain place, and the starting end point and the ending end point have been generated,

[0180] Step (3) is to cluster the end points of marking line and aggregate the center point. According to the end points of marking line {P i} generated in (2), the DBSCAN density clustering algorithm is used to cluster the end points of marking line, and the clustering cluster {C p} is obtained, wherein the eps neighborhood of each end point is obtained through the Euler distance d e (a, b) of two points, and the Euler distance d a (a, b) of point a (x a , y b ) and point b (x b , y e ) is obtained.(a, b), i.e. the straight-line distance of two points, is calculated as formula (1), the neighborhood radius is e, and the minimum number of points in the neighborhood of the core object is MinPts. For each cluster C p , the center point record is selected as its centroid, and the centroid is calculated as formula (2).

[0181]

[0182]

[0183]

[0184] (x i ,y i )∈C (2)

[0185] As shown in Figure 3 , for a core object p, its neighborhood eps is defined to include objects p1, p2, p3, p4, which are to-be-judged objects. In the cluster expansion, only the to-be-judged objects are expanded, and the construction of the cluster C p is completed.

[0186] Step (4) is: performing secondary clustering on the center points to obtain sub-road segment nodes, and sorting them according to the traffic flow direction. According to the center points generated in (2) and the directional road reference line read in (1), the projection operation is performed on the center points, and the center points are sorted according to the on-line distance d l (p, q) of the projection points. i Step (4) is: performing secondary clustering on the center points to obtain sub-road segment nodes, and sorting them according to the traffic flow direction. According to the center points generated in (2) and the directional road reference line read in (1), the projection operation is performed on the center points, and the center points are sorted according to the on-line distance d l (p, v s ) of the projection points p on the directional road reference line and the vector starting point v s , from small to large, and the serial number starts from 1.

[0187] Step (5) is: based on the sub-road segment nodes, the target marking line is broken, a new marking line and end point are formed, and the topological model is reconstructed. According to the sub-road segment nodes {BP i} generated in (4) and the target marking line-end point model generated in (2), the start and end points of the marking line and their associated sub-road segment node order are analyzed, and the target marking line whose start and end points are not adjacent is broken to form a new marking line-end point data. The breakpoint position is selected as the foot of the sub-road segment node on the target marking line.

[0188] As shown in Figure 4 , it is a reconstruction marking line-end point data diagram of the embodiment, through the breaking and supplementing of the end point and the marking line, the constraint condition that the start and end points of the marking line are adjacent is met, and the subsequent steps are met.

[0189] Step (6) is to cluster and sort the end points of the marking lines in the single sub-road segment node. According to the newly generated breakpoint-sub-road segment node relationship model in (5), the DBSCAN density clustering algorithm is used to cluster the end points in the sub-road segment node, wherein the eps neighborhood of each end point is determined by the Euclidean distance d e (a, b) is obtained, the neighborhood radius is e node , and the minimum number of points in the core object neighborhood is MinPts node . At the same time, the order of the clustering cluster is determined from small to large according to the distance of the clustering cluster center to the reference line of the directed road segment, wherein the distance on the left side of the line segment is negative, and the distance on the right side of the line segment is positive.

[0190] Step (7) is to extract adjacent marking lines and generate lanes based on the end point order in the sub-road segment node, and aggregate the lanes into lane groups based on the order of the sub-road segment node. According to the end point order in the sub-road segment node generated in (6), the target marking lines in the adjacent two sub-road segment nodes are sorted according to the end point order of the start and end points, and the lane lines between the adjacent marking lines are generated in sequence according to the sorting, as shown in Figure 5 , the shorter marking line of the left and right marking lines is selected as the main marking line L major , and the right marking line is selected when the lengths are the same, the other marking line is the reference line L refer , the perpendicular lines of each control point of L major are drawn to the reference line L refer in sequence, and the midpoints of the perpendicular line segments are selected as the control points of the lane lines, which are indented according to the predetermined rules and connected in sequence to generate the target lane lines, and the lane types are classified and the target road segment attribute information is corrected according to the marking line type information in the marking line data. At the same time, the lanes are aggregated into a series of lane groups in an upstream and downstream continuous manner according to the order relationship of the sub-road segment nodes. Define the lane data as structure Lane:

[0191] Lane={geom, type, Lline, Rline, Link, laneindex}

[0192] Wherein, geom is the lane line vector data, type is the lane line type information, Lline and Rline are left and right associated marking line information respectively, Link is the associated lane group information, and laneindex is the lane order.

[0193] Define the lane group data as structure Link:

[0194] Link={fbp, tbp, linkindex, R}

[0195] Wherein, fbp and tbp are associated start and end sub-road segment node information respectively, linkindex is the lane group order, and R is the associated directed road segment reference line.

[0196] Step (8) specifically involves supplementing lane connectors between lanes based on the topological relationships between lanes and lane groups. Based on the lane geometry and lane group topological relationships generated in (7), the traffic flow organization relationship between lanes is determined, and corresponding lane connectors are generated to ensure that the computable road network remains geometrically and logically connected. The geometry of the lane connectors is generated by quadratic and cubic Bézier curves. The calculation method for quadratic Bézier curves is shown in formula (3), and the calculation method for cubic Bézier curves is shown in formula (4).

[0197] LC = (1-t) 2 P0+2t(1-t)P0′+t 2 P1, t∈[0, 1] (3)

[0198] LC = (1-t) 3 P0+3t(1-t) 2 P0′+3t 2 (1-t)P1′+t 3 P1, t∈[0, 1] (4)

[0199] Wherein, P0 and P1 are the endpoints of the upstream and downstream lanes, and P0′ and P1′ are the selected control points.

[0200] like Figure 6 As shown, the upstream lane 1 and the downstream lane 2 can intersect at point P0′ through their extensions. The end point of lane 1 is P0, and the starting point of lane 2 is P1. Based on the spatial relationship, a lane connector (laneconnector) can be generated using a quadratic Bézier curve.

[0201] Finally Figure 7 As shown, a specific area in location A was selected as the implementation case. Through accurate labeling of high-precision road marking data and selection of appropriate directed road segment reference lines, a complete computable road network data model was successfully generated. In this model, lanes and lane connectors effectively reflect the drivable areas on the road and have complete logical and geometric topological relationships.

[0202] In summary, this invention can divide lane groups by clustering and sorting endpoints; it can determine adjacent road markings and generate lanes by clustering and sorting endpoints within sub-segment nodes; and it can determine the connectivity between lanes and generate lane connectors based on the topological relationships between lanes and lane groups. By annotating key road marking information, this invention successfully achieves the automatic editing and generation of lane-level road networks. The generated data meets the data structure requirements of lane-level road networks and matches the road traffic conditions reflected by the road marking data. This greatly improves the processing efficiency and operational accuracy of basic data, providing fundamental data support for the rapid development of intelligent transportation systems.

[0203] In some alternative embodiments, the function / operations mentioned in the block diagrams can not occur in the order mentioned in the operational illustrations. For example, two blocks shown in succession can in fact be executed substantially concurrently or the blocks can sometimes be executed in the reverse order, depending upon the functionality / operations involved. Also, embodiments presented and described in the flow diagrams are only examples of the method that can be employed in the present application. The disclosed method is not limited to the order of operations presented in the operational illustrations. Alternative embodiments are possible where the order of various operations is changed and where sub operations described as part of a larger operation are executed in a different order or are executed concurrently.

[0204] Moreover, while the present application has been described in the context of functional modules, it is to be understood that one or more of the functions and / or features described can be integrated in a single physical device and / or software module, or one or more functions and / or features can be implemented in separate physical devices or software modules. It will also be appreciated that detailed discussion of the actual implementation of each module is not necessary for an understanding of the application. Rather, the properties, functions and internal relationships of the various functional modules disclosed in the devices herein are deemed to be of a nature that would be understood by an engineer once presented with the property, function and internal relationships of the modules. Therefore, the engineer, using ordinary skill in the art, will be able to implement the application as claimed in the claims without undue experimentation. It will also be appreciated that the particular conceptualization disclosed is merely exemplary but the scope of the application is determined by the full breadth of the claims and equivalents thereof.

[0205] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the part of the technical solutions that make essential contributions to the prior art or the part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0206] The logic and / or steps represented in the flowcharts and / or described herein, for example, can be embodied in non-transitory computer-readable media, executed by one or more computing devices, and / or in any other way. The logic and / or steps represented in the flowcharts and / or described herein, for example, can be considered a list of executable instructions for implementing logic functions, and can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor- containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For purposes of this specification, a "computer-readable medium" can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The computer-readable medium can be, for example but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection (electronic) having one or more wires, a portable computer diskette (magnetic), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber (optical), and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example via an optical scanner, then compiled, interpreted, or otherwise processed, and stored in a computer memory in a form that can be later executed by a computer. In this context, a "computer-readable medium" can be any means that can store the program for use by or in connection with the instruction execution system, apparatus, or device.

[0207] The computer-readable medium can even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example via an optical scanner, then compiled, interpreted, or otherwise processed, and stored in a computer memory in a form that can be later executed by a computer.

[0208] It should be understood that aspects of the application can be implemented in hardware, software, firmware, or combinations thereof. In the embodiments described above, various steps or methods can be implemented, for example, using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, and in another embodiment, any of the following can be used: discrete logic circuitry, application specific integrated circuits (ASICs), programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), and / or the like.

[0209] In the description of the specification, the description using the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" etc. means that the specific feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In the specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific feature, structure, material, or characteristic described can be combined in any appropriate way in one or more embodiments or examples.

[0210] While the embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary and are not to be construed as limiting the scope of the application. The scope of the application is defined by the appended claims and their equivalents.

[0211] The above is the specific description of the preferred embodiment of the application, but the application is not limited to the described embodiments, and those skilled in the art can make various equivalent modifications or replacements without departing from the spirit of the application, and these equivalent modifications or replacements are all included in the scope defined by the claims of the application.

Claims

1. A lane-level high-precision computable road network construction method, characterized in that, The method comprises the following steps: acquiring lane marking data and a directional road section reference line, screening a target lane marking type, and grouping each lane marking in the lane marking data; adjusting the vector direction of the lane marking to generate end point information of the lane marking; wherein the end point information comprises start point information and end point information; performing first clustering processing on each end point of the lane marking to obtain a center point of the lane marking; performing second clustering processing on the center point to obtain a sub-road section node, and sorting the sub-road section node according to a traffic flow direction; after determining the target lane marking according to the target lane marking type, performing breaking processing on the target lane marking according to the sub-road section node to generate a new lane marking and corresponding end point information, and constructing a new breakpoint-sub-road section node topology model; performing clustering and sorting on the lane marking end points in a single sub-road section node; extracting adjacent lane markings and generating a lane according to the sorting result of the lane marking end points in the sub-road section node, and aggregating the lanes into a lane group based on the order of the sub-road section node; supplementing a lane connector between lanes according to the topological relationship between the lanes and the lane groups; According to the sub-road section node, the target mark line is interrupted to generate a new mark line and corresponding endpoint information, specifically: according to the sub-road section node and endpoint information, the start and end points of the mark line and the associated sub-road section node order are analyzed, the target mark line with non-adjacent start and end points is interrupted to form new mark line-end point data; wherein the breakpoint position is selected as the foot of the sub-road section node on the target mark line.

2. The lane-level high-precision computable road network construction method according to claim 1, characterized in that, in the step of acquiring lane marking data and a directional road section reference line, screening a target lane marking type, and grouping each lane marking in the lane marking data, the data structure of the lane marking data is as follows: wherein, represents the graticule data; is vector data; is graticule type information; is the associated directed road segment reference line; the target lane marking type comprises a crossable opposite lane boundary line, a crossable same-direction lane boundary line, a tidal lane line, a lane edge line, a harbor-type stop marking, a non-crossable opposite lane boundary line, a non-crossable same-direction lane boundary line, a bus lane line, and a variable guide lane marking; the data structure of the directional road section reference line is as follows: wherein, is a reference line for an associated directed road segment; is vector data.

3. The lane-level high-precision computable road network construction method according to claim 1, characterized in that, the step of adjusting the vector direction of the lane marking to generate end point information of the lane marking comprises: processing each directional road section and the associated lane marking according to the grouping condition of each lane marking in the lane marking data, adjusting the vector direction of the lane marking according to the traffic flow direction, generating lane end point information at the geometric start and end points, and constructing a corresponding topological association model; the data structure of the lane end point is as follows: wherein, the step of performing first clustering processing on each end point of the lane marking to obtain a center point of the lane marking comprises: reticle end point; , longitude and latitude of the end point; start and end point attribute identification of the end point; reticle information associated with the end point.

4. The lane-level high-precision computable road network construction method according to claim 1, characterized in that, performing clustering on the lane end points and aggregating the center points; performing clustering on the lane end points by using a DBSCAN density clustering algorithm according to the end point information of the lane marking to obtain a clustering cluster; wherein the neighborhood between each end point is obtained by using the Euler distance between two points; for each clustering cluster, selecting the centroid of the clustering cluster as a center point for recording. the step of performing second clustering processing on the center point to obtain a sub-road section node, and sorting the sub-road section node according to a traffic flow direction comprises:

5. The lane-level high-precision computable road network construction method according to claim 1, characterized in that, performing a projection operation on the center point according to the directional road section reference line and the center point of the lane marking; performing secondary clustering on the center point according to the on-line distance of the projection point to obtain a sub-road section node set; sequentially marking the center points from small to large according to the on-line distance between the projection point on the directional road section reference line and the vector start point, wherein the sequence number of the sorting starts from 1. the step of performing clustering and sorting on the lane marking end points in a single sub-road section node comprises:

6. The lane-level high-precision computable road network construction method according to claim 1, characterized in that, ​ According to the newly constructed breakpoint-subsection node topology model, the end points in the subsection node are clustered by using the DBSCAN density clustering algorithm; wherein the neighborhood of each end point is determined by the Euler distance of two points; According to the distance between the cluster center and the reference line of the directed subsection, the order of the cluster is determined from small to large; wherein the distance on the left side of the line segment is negative, and the distance on the right side of the line segment is positive.

7. The lane-level high-precision computable road network construction method according to claim 1, characterized in that, According to the sorting result of the marking line end points in the subsection node, the adjacent marking lines are extracted and the lanes are generated, and the lanes are aggregated into lane groups based on the order of the subsection node, including: According to the sorting result of the marking line end points in the subsection node, the target marking lines in the adjacent two subsection nodes are sorted according to the end point order of the start and end points; According to the sorting, the lane lines between the adjacent marking lines are generated in sequence; According to the order relationship of the subsection nodes, the lanes are aggregated into a series of upstream and downstream continuous lane groups; According to the sorting, the lane lines between the adjacent marking lines are generated in sequence, including: Selecting the shorter marking line of the left and right marking lines as the main marking line, selecting the right marking line as the main marking line when the left and right marking lines are the same length, selecting the other marking line outside the main marking line as the reference line, drawing a perpendicular line from each control point of the main marking line to the reference line in sequence, and selecting the midpoint of the perpendicular line segment as the control point of the lane line, and then performing indentation according to the predetermined rule, sequentially connecting to generate the target lane line, and classifying and correcting the target lane type according to the marking line type information in the marking line data.

8. A lane-level high-precision computable road network construction apparatus, characterized by comprising: Including: The first module is used for acquiring marking line data and directed subsection reference line, screening target marking line type, and grouping each marking line in the marking line data; The second module is used for adjusting the vector direction of the marking line and generating the end point information of the marking line; wherein the end point information includes start point information and end point information; The third module is used for first clustering processing of each end point of the marking line to obtain the center point of the marking line; The fourth module is used for second clustering processing of the center point to obtain the subsection node, and sorting the subsection node according to the traffic flow direction; The fifth module is used for determining the target marking line according to the target marking line type, performing breaking processing on the target marking line according to the subsection node to generate new marking lines and corresponding end point information, and constructing a new breakpoint-subsection node topology model; The sixth module is used for clustering and sorting the marking line end points in a single subsection node; The seventh module is used for extracting adjacent marking lines and generating lanes according to the sorting result of the marking line end points in the subsection node, and aggregating the lanes into lane groups based on the order of the subsection node; The eighth module is used for supplementing the lane connectors between the lanes according to the topological relationship between the lanes and the lane groups. According to the sub-road section node, the target mark line is interrupted to generate a new mark line and corresponding endpoint information, specifically: according to the sub-road section node and endpoint information, the start and end points of the mark line and the order of the associated sub-road section nodes are analyzed, the target mark line with non-adjacent start and end points is interrupted to form new mark line-end point data; wherein the breakpoint position is selected as the foot of the sub-road section node on the target mark line.

9. An electronic device, comprising: The processor and the memory are included; The memory is used for storing programs; The processor executes the programs to realize the method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The storage medium stores programs, and the programs are executed by the processor to realize the method of any one of claims 1-7.

Images