Method and device for generating virtual lanes in high-precision map intersections

Abstract

The present invention discloses a method and device for generating virtual lanes in a high-precision map intersection, including: obtaining the topological relationship between the intersection and its corresponding entry lane and exit lane; generating a regular grid of the intersection range based on the boundary of the intersection, and determining the passability of each grid in the regular grid; traversing each entry lane of the intersection, determining the passability of each exit lane of the intersection, and finding all passable paths in the intersection; calculating the lane centerline of each passable path using a path planning algorithm according to the regular grid; optimizing the lane centerline of each passable path obtained; generating the lane boundary line of the virtual lane in the intersection according to the optimized lane centerline of each passable path. The present invention can realize the automatic generation of virtual lanes in high-precision map intersections.

Description

Technical Field

[0001] The present invention belongs to the technical field of high-precision maps, and in particular relates to a method and device for generating a virtual lane in a high-precision map intersection. Background Art

[0002] As the infrastructure of the autonomous driving system, the precision and accuracy of high-precision maps are crucial to the autonomous driving system. The lane boundary lines on general roads can be obtained through the perception system and represented by printed lines, curbs, guardrails and isolation belts. The lane center line can be generated from the lane boundary lines. However, the traffic conditions at the intersection are complicated, and only the left turn waiting area line and the intersection guide line are available for reference. There may also be obstacles such as flower beds, isolation belts and no-driving zones at the intersection that need to be avoided, and there are not enough printed lines to rely on, which makes it difficult to automatically generate virtual lanes at the intersection. Summary of the invention

[0003] In order to solve the above technical problems, the present invention proposes a method and device for generating virtual lanes in a high-precision map intersection.

[0004] In order to achieve the above object, the technical solution of the present invention is as follows:

[0005] On the one hand, the present invention discloses a method for generating a virtual lane in a high-precision map intersection, comprising:

[0006] Step S1: Obtain the topological relationship between the intersection and its corresponding entry lane and exit lane;

[0007] Step S2: Based on the boundary of the intersection, a regular grid of the intersection range is generated, and the passability of each grid in the regular grid is determined;

[0008] Step S3: traverse each entrance lane of the intersection, determine the passability of each exit lane of the intersection, and find all passable paths in the intersection;

[0009] Step S5: Calculate the lane centerline of each passable path using a path planning algorithm based on the regular grid;

[0010] Step S6: Optimizing the lane centerline of each passable path obtained in step S5;

[0011] Step S7: Generate lane boundary lines of virtual lanes in the intersection based on the optimized lane center lines of each drivable path.

[0012] On the basis of the above technical solution, the following improvements can be made:

[0013] As a preferred solution, step S2 specifically includes:

[0014] Step S2.1: Determine the boundaries of the geometric surface elements of the intersection;

[0015] Step S2.2: Taking the boundary of the intersection as the grid range, generating a regular grid within the intersection range;

[0016] Step S2.3: Formulate traffic constraints;

[0017] Step S2.4: Based on the access constraint conditions, determine the accessibility of each grid in the regular grid.

[0018] As a preferred solution, step S5 specifically includes:

[0019] Step S5.1: For each passable path, determine the grid positions of the starting point and the end point in the path planning algorithm under the regular grid by the positions of the end point of the centerline of the entry lane and the beginning point of the centerline of the exit lane;

[0020] Step S5.2: Based on the grid positions of the start and end points of each traversable path, a path planning algorithm is used to obtain the lane centerline of each traversable path.

[0021] As a preferred solution, step S6 specifically includes:

[0022] Step S6.1: Summarize a number of different problems existing in the lane centerline of each passable path, wherein: the problems include: the geometry of the lane centerline is jagged, the geometry of the lane centerline does not meet the dynamic requirements, and the connection between the lane centerline and the centerline of the entry lane and the exit lane is not smooth;

[0023] Step S6.2: For different problems, select the corresponding algorithm to optimize the lane centerline to obtain a smooth curved lane centerline.

[0024] As a preferred solution, between step S3 and step S5, the following is further included:

[0025] Step S4: Verify each passable path, including:

[0026] Step S4.1: Obtain the driving direction of the drivable path according to the geometric information of the tail section of the centerline of the entry lane and the first section of the centerline of the exit lane;

[0027] Step S4.2: Obtain the direction of travel of the entrance lane of the intersection according to the guide arrow of the intersection;

[0028] Step S4.3: Verify whether the driving direction of the passable path obtained in step S4.1 is consistent with the driving direction of the entry lane obtained in step S4.2;

[0029] If they are consistent, the pass verification of the passable path is qualified;

[0030] Otherwise, failed.

[0031] In addition, on the other hand, the present invention also discloses a device for generating a virtual lane in a high-precision map intersection, comprising:

[0032] A topological relationship acquisition module is used to acquire the topological relationship between the intersection and its corresponding entry lane and exit lane;

[0033] A regular grid generation module is used to generate a regular grid within the intersection range based on the boundary of the intersection and determine the passability of each grid in the regular grid;

[0034] The passable path search module is used to traverse each entrance lane of the intersection, determine the passability of each exit lane of the intersection, and find all passable paths in the intersection;

[0035] The lane centerline calculation module is used to calculate the lane centerline of each passable path based on the regular grid using the path planning algorithm;

[0036] A lane centerline optimization module, used to optimize the lane centerline of each passable path obtained by the lane centerline calculation module;

[0037] The lane boundary line generation module is used to generate lane boundary lines of virtual lanes in the intersection according to the optimized lane center lines of each passable path.

[0038] As a preferred solution, the regular grid generation module specifically includes:

[0039] A boundary determination unit, used to determine the boundary of the geometric surface element of the intersection;

[0040] A regular grid generation unit, used to generate a regular grid within the intersection range by taking the boundary of the intersection as the grid range;

[0041] A constraint condition formulation unit, used to formulate traffic constraint conditions;

[0042] The passability determination unit is used to determine the passability of each grid in the regular grid based on the passability constraint condition.

[0043] As a preferred solution, the lane centerline calculation module specifically includes:

[0044] A starting point and end point determination unit is used to determine the grid positions of the starting point and the end point in the path planning algorithm under a regular grid by using the positions of the end point of the center line of the entry lane and the beginning point of the center line of the exit lane for each passable path;

[0045] The lane centerline calculation unit is used to obtain the lane centerline of each passable path using a path planning algorithm according to the grid positions of the starting point and the end point of each passable path.

[0046] As a preferred solution, the lane centerline optimization module specifically includes:

[0047] A problem summarizing unit is used to summarize a number of different problems existing in the lane centerline of each passable path, wherein: the problems include: the geometric shape of the lane centerline has jagged edges, the geometric shape of the lane centerline meets the dynamic requirements of a broken line, and the connection between the lane centerline and the centerline of the entry lane and the exit lane is not smooth.

[0048] The optimization unit is used to select corresponding algorithms for optimizing the lane center line for different problems, so as to obtain a smooth curved lane center line.

[0049] As a preferred solution, it also includes: a verification module, which is used to verify the passability of each passable path;

[0050] The verification modules include:

[0051] A driving direction acquisition unit, used to obtain the driving direction of the passable path according to the geometric information of the tail section of the center line of the entry lane and the first section of the center line of the exit lane;

[0052] A traffic direction acquisition unit, used to acquire the traffic direction of the entry lane of the intersection according to the guide arrow of the intersection;

[0053] A verification unit, used to verify whether the driving direction of the passable path obtained by the driving direction acquisition unit is consistent with the passing direction of the entering lane obtained by the passing direction acquisition unit;

[0054] If they are consistent, the pass verification of the passable path is qualified;

[0055] Otherwise, unqualified.

[0056] The present invention discloses a method and device for generating a virtual lane in a high-precision map intersection, which has the following beneficial effects:

[0057] The present invention can obtain a smooth curved lane centerline, and based on the lane centerline, a lane boundary line of a virtual lane in an intersection can be generated, thereby realizing the automatic generation of a virtual lane in a high-precision map intersection. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0059] Figure 1 A flowchart of a generation method provided by an embodiment of the present invention.

[0060] Figure 2 A schematic diagram of the topological relationship between an intersection and its corresponding entry lanes and exit lanes provided in an embodiment of the present invention.

[0061] Figure 3 A schematic diagram of a regular grid at an intersection provided in an embodiment of the present invention.

[0062] Figure 4 A schematic diagram of lane travel direction provided by an embodiment of the present invention.

[0063] Figure 5 The lane centerline obtained based on the original A* algorithm provided in the embodiment of the present invention.

[0064] Figure 6 A comparison diagram of the lane centerline jagged condition before and after optimization provided in an embodiment of the present invention.

[0065] Figure 7 A comparison diagram of the turning point of the lane centerline before and after optimization provided in an embodiment of the present invention.

[0066] Figure 8 A comparison diagram of the lane centerline connection before and after optimization provided in an embodiment of the present invention.

[0067] Fig. 9 A schematic diagram of generating lane boundary lines within an intersection provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0068] The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0069] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] In addition, the expression of “comprising” an element is an “open” expression, which merely means that corresponding components or steps exist, and should not be interpreted as excluding additional components or steps.

[0071] In order to achieve the purpose of the present invention, a method and device for generating a virtual lane in a high-precision map intersection in some embodiments, such as Figure 1 As shown, the generation method includes:

[0072] Step S1: Obtain the topological relationship between the intersection and its corresponding entry lane and exit lane;

[0073] Step S2: Based on the boundary of the intersection, a regular grid of the intersection range is generated, and the passability of each grid in the regular grid is determined;

[0074] Step S3: traverse each entrance lane of the intersection, determine the passability of each exit lane of the intersection, and find all passable paths in the intersection;

[0075] Step S4: Verify each passable path;

[0076] Step S5: Calculate the lane centerline of each verified passable path using a path planning algorithm according to the regular grid;

[0077] Step S6: Optimizing the lane centerline of each passable path obtained in step S5;

[0078] Step S7: Generate lane boundary lines of virtual lanes in the intersection based on the optimized lane center lines of each drivable path.

[0079] The present invention can automatically generate virtual lanes in intersections quickly and accurately.

[0080] Each step is described in detail below.

[0081] Step S1: Obtain the topological relationship between the intersection and its corresponding entry lane and exit lane, specifically:

[0082] For each intersection, find the lanes that are adjacent to it, determine all the entry and exit lanes associated with the intersection, bind the boundary lines and center lines of each entry and exit lane to the intersection, and obtain the corresponding topological relationship, such as Figure 2 shown.

[0083] Step S2: Based on the boundary of the intersection, a regular grid of the intersection range is generated, and the passability of each grid in the regular grid is determined, such as Figure 3 As shown, specifically including:

[0084] Step S2.1: Determine the boundary of the geometric surface element of the intersection. The scope of the intersection is generally described using geometric surface elements;

[0085] Step S2.2: Taking the boundary of the intersection as the grid range, a regular grid of the intersection range is generated, and each grid in the regular grid is generally a square;

[0086] Step S2.3: Establish traffic restriction conditions, as follows:

[0087] (1) The grid outside the intersection geometry is inaccessible;

[0088] (2) The grids where curbstones, median strips, no-driving zones, flower beds and solid printed lines pass through the intersection are not accessible;

[0089] (3) The grid within 0.5 m from the actual boundary line, curb and solid line of the intersection is not passable;

[0090] Step S2.4: Based on the access constraint conditions, determine the accessibility of each grid in the regular grid.

[0091] Step S3: traverse each entrance lane of the intersection, determine the passability of each exit lane of the intersection, and find all passable paths in the intersection.

[0092] Step S4: Verify each passable path, including:

[0093] Step S4.1: According to the geometric information of the end of the centerline of the incoming lane and the beginning of the centerline of the outgoing lane, the driving direction of the passable path is obtained, such as: straight, left or right turn, and U-turn;

[0094] Step S4.2: Obtain the direction of travel of the entrance lane of the intersection according to the guide arrow of the intersection;

[0095] Step S4.3: Verify whether the driving direction of the passable path obtained in step S4.1 is consistent with the driving direction of the entry lane obtained in step S4.2;

[0096] If they are consistent, the pass verification of the passable path is qualified, and the path is determined to be passable;

[0097] Otherwise, unqualified.

[0098] In step S4.2, for each lane entering the intersection, the lane's travel direction attribute can be determined by the ground guide arrow at the intersection in the lane, such as Figure 4 Generally, the first set of guide arrows closest to the intersection is 3m to 5m away from the stop line, and the second set of guide arrows is at the starting position of the guide lane.

[0099] For each incoming lane, you can trace back from the stop line of the intersection to the upstream of the lane, bind the ground guide arrow in the lane to the lane, and determine the lane's travel direction information based on the guide arrow and traffic regulations.

[0100] Step S5: Based on the regular grid, the path planning algorithm is used to calculate the lane centerline of each qualified passable path, specifically including:

[0101] Step S5.1: For each passable path, determine the grid positions of the starting point and the end point in the path planning algorithm under the regular grid by the positions of the end point of the centerline of the entry lane and the beginning point of the centerline of the exit lane;

[0102] Step S5.2: Based on the grid positions of the start and end points of each traversable path, a path planning algorithm is used to obtain the lane centerline of each traversable path.

[0103] Wherein: the above path planning algorithm can be but is not limited to the A* algorithm.

[0104] The center line of the lane in the intersection obtained based on the original A* algorithm is as follows Figure 5 As shown, its geometric shape does not meet the requirements and needs to be optimized later.

[0105] Step S6: Optimizing the lane centerline of each passable path obtained in step S5, specifically including:

[0106] Step S6.1: Summarize a number of different problems existing in the lane centerline of each passable path, wherein: the problems include: the geometry of the lane centerline is jagged, the geometry of the lane centerline is a broken line that does not meet the dynamic requirements, and the connection between the lane centerline and the centerlines of the entry lane and the exit lane is not smooth;

[0107] Step S6.2: For different problems, select the corresponding algorithm to optimize the lane centerline to obtain a smooth curved lane centerline.

[0108] Furthermore, the lane centerline generated by the original A* algorithm has some flaws, mainly in the following aspects:

[0109] (1) The geometry of the generated lane centerline is jagged;

[0110] (2) The geometry of the generated lane centerline is a broken line, which does not meet the dynamic requirements;

[0111] (3) The junction between the generated lane centerline and the centerlines of the entry and exit lanes is not smooth enough.

[0112] To address the above three issues, the following three algorithms can be used to optimize the geometry of the lane centerline within the intersection generated based on the original A* algorithm.

[0113] For question (1): Figure 6 As shown in (a), the geometry of the generated lane centerline is jagged and is optimized using two processes.

[0114] In process 1, the lane centerline is generated using the publicly available optimized A* algorithm from “Path planning and smoothing of mobile robot based on improved artificial fish swarm algorithm”.

[0115] In the classic grid-based A* algorithm, the step length of each movement is 1, and the moving directions are east, south, west, and north. The distance is calculated in the heuristic function as Manhattan distance d = |x1-x2|+|y1-y2|.

[0116] In order to reduce the occurrence of aliasing, the above-mentioned optimized A* algorithm is expanded from the original eight directions of movement (up, down, left, right, up, left, down, right) to 16 directions and 24 neighborhoods (you can imagine it as a 5*5 grid starting from the center grid and being able to move to any other grid), and the way to calculate the distance in the heuristic function is adjusted to Euclidean distance. After the above adjustments, the classic grid-based A* algorithm will generate a less jagged path.

[0117] Process 2: post-process the lane centerline generated in process 1, optimize it using the line of sight algorithm, and “straighten” the jagged shape.

[0118] After the above two processes, a path without jagged edges can be obtained, such as Figure 6 (b) as shown.

[0119] For question (2): Figure 7 As shown in (a), the geometric shape of the lane centerline generated is a broken line, which does not meet the dynamic requirements. Since the path obtained by the grid-based A* path planning algorithm is a broken line, the turning point is abrupt, which does not meet the mapping requirements and needs to be optimized. In this case, the lane centerline obtained by the path planning algorithm is processed by Bezier interpolation, that is, a smooth curve can be obtained, as shown in Figure 7 (b) as shown.

[0120] For question (3): Figure 8As shown in (a), the generated lane centerline is not smooth enough at the junction with the centerline of the entry lane and the exit lane. In this case, the optimization is performed by extending the boundary lines of the entry lane and the exit lane. The schematic diagram of the lane centerline geometry optimization is shown in Figure 8 (b) as shown.

[0121] Step S7: Generate lane boundary lines of virtual lanes in the intersection according to the optimized lane center lines of each drivable path, specifically including:

[0122] Step S7.1: Generate two parallel lines on both sides of the lane centerline in the intersection as initial lane boundary lines. The distance between the parallel lines and the lane centerline can be set to a certain value (eg, 1.5 m).

[0123] Step S7.2: Select the end points of the left and right boundary lines of the upstream lane and the beginning points of the left and right boundary lines of the downstream lane of the lane centerline in the intersection as control points, and adjust the coordinates of the parallel lines generated in step S7.1 in proportion to the distance to match the upstream and downstream boundary lines;

[0124] Step S7.3: Generate a virtual lane in the intersection, such as Fig. 9 shown.

[0125] Through the above steps S1 to S7, the geometric elements of the lanes in the intersection can be obtained, and then the attribute values ​​are added thereto to obtain the virtual lanes in the intersection.

[0126] In addition, an embodiment of the present invention further discloses a device for generating a virtual lane in a high-precision map intersection, comprising:

[0127] A topological relationship acquisition module is used to acquire the topological relationship between the intersection and its corresponding entry lane and exit lane;

[0128] A regular grid generation module is used to generate a regular grid within the intersection range based on the boundary of the intersection and determine the passability of each grid in the regular grid;

[0129] The passable path search module is used to traverse each entrance lane of the intersection, determine the passability of each exit lane of the intersection, and find all passable paths in the intersection;

[0130] A verification module, used to verify the passability of each passable path;

[0131] The lane centerline calculation module is used to calculate the lane centerline of each verified passable path based on the regular grid using the path planning algorithm;

[0132] A lane centerline optimization module, used to optimize the lane centerline of each passable path obtained by the lane centerline calculation module;

[0133] The lane boundary line generation module is used to generate lane boundary lines of virtual lanes in the intersection according to the optimized lane center lines of each passable path.

[0134] Furthermore, the regular grid generation module specifically includes:

[0135] A boundary determination unit, used to determine the boundary of the geometric surface element of the intersection;

[0136] A regular grid generation unit, used to generate a regular grid within the intersection range by taking the boundary of the intersection as the grid range;

[0137] A constraint condition formulation unit, used to formulate traffic constraint conditions;

[0138] The passability determination unit is used to determine the passability of each grid in the regular grid based on the passability constraint condition.

[0139] Furthermore, the lane centerline calculation module specifically includes:

[0140] A starting point and end point determination unit is used to determine the grid positions of the starting point and the end point in the path planning algorithm under a regular grid by using the positions of the end point of the center line of the entry lane and the beginning point of the center line of the exit lane for each passable path;

[0141] The lane centerline calculation unit is used to obtain the lane centerline of each passable path using a path planning algorithm according to the grid positions of the starting point and the end point of each passable path.

[0142] Furthermore, the lane centerline optimization module specifically includes:

[0143] A problem summarizing unit is used to summarize a number of different problems existing in the lane centerline of each passable path, wherein: the problems include: the geometric shape of the lane centerline has jagged edges, the geometric shape of the lane centerline is a broken line that does not meet the dynamic requirements, and the connection between the lane centerline and the centerline of the entry lane and the exit lane is not smooth.

[0144] The optimization unit is used to select corresponding algorithms for optimizing the lane center line for different problems, so as to obtain a smooth curved lane center line.

[0145] Furthermore, the verification module includes:

[0146] A driving direction acquisition unit, used to obtain the driving direction of the passable path according to the geometric information of the tail section of the center line of the entry lane and the first section of the center line of the exit lane;

[0147] A traffic direction acquisition unit, used to acquire the traffic direction of the entry lane of the intersection according to the guide arrow of the intersection;

[0148] A verification unit, used to verify whether the driving direction of the passable path obtained by the driving direction acquisition unit is consistent with the passing direction of the entering lane obtained by the passing direction acquisition unit;

[0149] If they are consistent, the pass verification of the passable path is qualified;

[0150] Otherwise, failed.

[0151] The device for generating virtual lanes in high-precision map intersections disclosed in the present invention is similar to the method for generating the virtual lanes, which will not be described in detail here.

[0152] The present invention discloses a method and device for generating a virtual lane in a high-precision map intersection, which has the following beneficial effects:

[0153] The present invention can obtain a smooth curved lane centerline, and based on the lane centerline, a lane boundary line of a virtual lane in an intersection can be generated, thereby realizing the automatic generation of a virtual lane in a high-precision map intersection.

[0154] It should be understood that the various techniques described herein can be implemented in combination with hardware or software, or a combination thereof. Thus, the method and apparatus of the present invention, or some aspects or portions of the method and apparatus of the present invention, can take the form of program codes (i.e., instructions) embedded in a tangible medium, such as a floppy disk, a CD-ROM, a hard disk drive, or any other machine-readable storage medium, wherein when the program is loaded into a machine such as a computer and executed by the machine, the machine becomes a device for practicing the present invention.

[0155] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable ordinary technicians in the field to understand the content of the present invention and implement it. They cannot be used to limit the protection scope of the present invention. All equivalent changes or modifications made according to the spirit of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for generating a virtual lane in a high-precision map intersection, characterized in that: include: Step S1: Obtain the topological relationship between the intersection and its corresponding entry lane and exit lane; Step S2: Based on the boundary of the intersection, a regular grid of the intersection range is generated, and the passability of each grid in the regular grid is determined; Step S3: traverse each entrance lane of the intersection, determine the passability of each exit lane of the intersection, and find all passable paths in the intersection; Step S5: Calculate the lane centerline of each passable path using a path planning algorithm based on the regular grid; Step S6: Optimizing the lane centerline of each passable path obtained in step S5; Step S7: Generate lane boundary lines of virtual lanes in the intersection based on the optimized lane center lines of each drivable path.

2. The generation method according to claim 1, characterized in that: The step S2 specifically includes: Step S2.1: Determine the boundaries of the geometric surface elements of the intersection; Step S2.2: Taking the boundary of the intersection as the grid range, generating a regular grid within the intersection range; Step S2.3: Formulate traffic constraints; Step S2.4: Based on the access constraint conditions, determine the accessibility of each grid in the regular grid.

3. The generation method according to claim 1, characterized in that: The step S5 specifically includes: Step S5.1: For each passable path, determine the grid positions of the starting point and the end point in the path planning algorithm under the regular grid by the positions of the end point of the centerline of the entry lane and the beginning point of the centerline of the exit lane; Step S5.2: Based on the grid positions of the start and end points of each traversable path, a path planning algorithm is used to obtain the lane centerline of each traversable path.

4. The generation method according to claim 1, characterized in that: The step S6 specifically includes: Step S6.1: Summarize a number of different problems existing in the lane centerline of each passable path, wherein: the problems include: the geometry of the lane centerline is jagged, the geometry of the lane centerline does not meet the dynamic requirements, and the connection between the lane centerline and the centerline of the entry lane and the exit lane is not smooth; Step S6.2: For different problems, select the corresponding algorithm to optimize the lane centerline to obtain a smooth curved lane centerline.

5. The generation method according to any one of claims 1 to 4, characterized in that: The steps S3 and S5 also include: Step S4: Verify each passable path, including: Step S4.1: Obtain the driving direction of the drivable path according to the geometric information of the tail section of the centerline of the entry lane and the first section of the centerline of the exit lane; Step S4.2: Obtain the direction of travel of the entrance lane of the intersection according to the guide arrow of the intersection; Step S4.3: Verify whether the driving direction of the passable path obtained in step S4.1 is consistent with the driving direction of the entry lane obtained in step S4.2; If they are consistent, the pass verification of the passable path is qualified; Otherwise, failed.

6. A device for generating virtual lanes in high-precision map intersections, characterized in that: include: A topological relationship acquisition module is used to acquire the topological relationship between the intersection and its corresponding entry lane and exit lane; A regular grid generation module is used to generate a regular grid within the intersection range based on the boundary of the intersection and determine the passability of each grid in the regular grid; The passable path search module is used to traverse each entrance lane of the intersection, determine the passability of each exit lane of the intersection, and find all passable paths in the intersection; The lane centerline calculation module is used to calculate the lane centerline of each passable path based on the regular grid using the path planning algorithm; A lane centerline optimization module, used to optimize the lane centerline of each passable path obtained by the lane centerline calculation module; The lane boundary line generation module is used to generate lane boundary lines of virtual lanes in the intersection according to the optimized lane center lines of each passable path.

7. The generating device according to claim 6, characterized in that The regular grid generation module specifically includes: A boundary determination unit, used to determine the boundary of the geometric surface element of the intersection; A regular grid generation unit, used to generate a regular grid within the intersection range by taking the boundary of the intersection as the grid range; A constraint condition formulation unit, used to formulate traffic constraint conditions; The passability determination unit is used to determine the passability of each grid in the regular grid based on the passability constraint condition.

8. The generating device according to claim 6, characterized in that The lane centerline calculation module specifically includes: A starting point and end point determination unit is used to determine the grid positions of the starting point and the end point in the path planning algorithm under a regular grid by using the positions of the end point of the center line of the entry lane and the beginning point of the center line of the exit lane for each passable path; The lane centerline calculation unit is used to obtain the lane centerline of each passable path using a path planning algorithm according to the grid positions of the starting point and the end point of each passable path.

9. The generating device according to claim 6, characterized in that The lane centerline optimization module specifically includes: A problem summarizing unit is used to summarize a number of different problems existing in the lane centerline of each passable path, wherein: the problems include: the geometric shape of the lane centerline has jagged edges, the geometric shape of the lane centerline does not meet the dynamic requirements, and the connection between the lane centerline and the centerline of the entry lane and the exit lane is not smooth. The optimization unit is used to select corresponding algorithms for optimizing the lane center line for different problems, so as to obtain a smooth curved lane center line.

10. The generating device according to any one of claims 6 to 9, characterized in that: Also includes: A verification module, used to verify the passability of each passable path; The verification module comprises: A driving direction acquisition unit, used to obtain the driving direction of the passable path according to the geometric information of the tail section of the center line of the entry lane and the first section of the center line of the exit lane; A traffic direction acquisition unit, used to acquire the traffic direction of the entry lane of the intersection according to the guide arrow of the intersection; A verification unit, used to verify whether the driving direction of the passable path obtained by the driving direction acquisition unit is consistent with the passing direction of the entering lane obtained by the passing direction acquisition unit; If they are consistent, the pass verification of the passable path is qualified; Otherwise, failed.

Images