An incremental Voronoi road network construction method

Through the incremental Vino Road Network construction method, the problem of insufficient adaptability of Vino graph construction based on line and surfaces in the existing technology is solved, and efficient Road network update and path planning.

CN115063548BActive Publication Date: 2024-11-29SOUTHEAST UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210676432.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-06-15
Publication Date
2024-11-29
Estimated Expiration
2042-06-15

AI Technical Summary

Technical Problem

Existing algorithms are difficult to directly construct line-plane-based Vino graphs, and lack the ability to re-plan maps to adapt to new obstacles, resulting in waste of resources and inefficiency.

Method used

The incremental Vino road network construction method is adopted, and the connecting domains in the map are numbered and triangulated, and the road network nodes are built by sets of discrete edge points to adapt to new obstacles and integrate local road networks with global road networks.

Benefits of technology

It improves algorithm efficiency, reduces the number of discrete points, can adapt to new obstacles without destroying the global road network, achieve seamless integration of local road network and global road network, and provides fast path indexing and path planning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115063548B_ABST
    Figure CN115063548B_ABST
Patent Text Reader

Abstract

The present invention discloses an incremental Voronoi road network construction method. The method comprises: firstly, numbering obstacles in a map, extracting edges, centroids, and circumscribed frames, triangulating the image using the centroids, pairing obstacles according to common edge relationships, discretizing obstacle edges, constructing a Voronoi diagram, obtaining road network nodes, and establishing adjacency relationships between road network nodes; secondly, when a new obstacle appears in the map, extracting the edge, centroid, and circumscribed frame of the new obstacle, judging the impact of the new obstacle on the original map road network, constructing a local map, and generating a local road network; finally, fusing the global and local road networks. The present invention can construct a Voronoi road network based on line-surface generators, and locally update the road network when a new obstacle is detected, and seamlessly merge it with the global road network, thereby improving the efficiency of the algorithm.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to an incremental Voronoi road network construction method, belongs to a path planning technology, and is specifically applicable to the field of road network extraction. Background Art

[0002] The road network model needs to be able to reflect the characteristics of the road network truthfully and support various application requirements for road network constraints, such as trajectory indexing and trajectory query. In geographic information systems, road network models are often defined as undirected or directed network topology graphs, which can be intuitively represented as a ground plane network graph composed of several points and lines. The Voronoi diagram is a seamless segmentation grid diagram in space. Due to the particularity of its data structure, it can be applied to the segmentation of neighboring areas in two-dimensional space and can be applied to road network extraction.

[0003] Obstacles in maps usually exist in the form of lines or surfaces. Existing algorithms are very effective in constructing Voronoi diagrams of planar point sets, but are difficult to be directly used to construct Voronoi diagrams based on lines and surfaces. They also lack the ability to adapt to new obstacles generated in the map, and replanning the global road network will consume a lot of resources. In order to construct a Voronoi road network based on an obstacle map and adapt to new obstacles, it is urgent to propose an incremental Voronoi road network construction method. Summary of the invention

[0004] In view of the above problems, the present invention provides an incremental Voronoi network construction method.

[0005] To achieve the above purpose, the technical solution of the present invention is as follows:

[0006] An incremental Voronoi road network construction method comprises the following steps:

[0007] S1: Number the obstacles based on the connected domains in the map, and obtain the edge, centroid and external frame of the connected domains;

[0008] S2: triangulate the image using the centroid of the connected domain and pair the regions according to the three-edge relationship;

[0009] S3: Discretize the edges of each connected domain, construct a Voronoi graph based on the discrete edge point set, and obtain a road network node set consisting of Voronoi vertices;

[0010] S4: Associating each road network node with its adjacent point according to the vertex common edge relationship to obtain the original map Voronoi road network;

[0011] S5: Compare the new map with the original map to obtain the coordinates of the new obstacle, and extract the edge, centroid, and external frame of the connected domain formed by the new obstacle;

[0012] S6: Determine whether the original map road network crosses the external frame formed by the new obstacle. If so, proceed to step S7; if not, do not process;

[0013] S7: Taking the centroid of the new obstacle as the center, the local map is gradually expanded by the multiple of the length and width of the external frame until there are three or more connected domains in the local map, and the original road network nodes in the local map are removed;

[0014] S8: Repeat steps S3 and S4 for the local map to obtain local road network nodes;

[0015] S9: Merge the road network in the local map with the road network in the original map.

[0016] Preferably, the step S1 specifically includes the following process:

[0017] Preprocess the imported binary map: use functions to distinguish each connected domain, extract the edge, centroid and external frame coordinates of each connected domain, and concatenate the centroid coordinates into an array.

[0018] Preferably, the step S2 specifically includes the following process:

[0019] The image is Delaunay triangulated using the centroid coordinate array. The triangle vertices correspond to the numbers of three adjacent connected domains. The connected domains are paired according to the triangle co-edge relationship. The connected domains at both ends of the same edge are a pair.

[0020] Preferably, the step S3 specifically includes the following process:

[0021] (3-1) The edges of each connected domain are discretized, and the discrete step length is related to the distance between the paired elements, expressed as:

[0022]

[0023] Among them, d is the shortest distance between paired elements, L1 and L2 are the side length sets of paired elements, and λ is the discrete step length.

[0024] (3-2) Taking the discrete edge point set as the generating point set, generate the Voronoi diagram and Voronoi vertex coordinates;

[0025] (3-3) Merge the Voronoi regions generated by the same connected domain, that is, remove redundant vertices, including vertices outside the map and vertices within the connected region. If the line segment generated by two adjacent vertices crosses the connected region, the two vertices are removed, and the remaining vertices constitute the road network nodes in the map, which are numbered.

[0026] Preferably, the step S4 specifically includes the following process:

[0027] Each road network node is associated with its neighboring points according to the common edge relationship of the original Voronoi vertices, and the adjacent point number, coordinates, and distance of each node are recorded, which can be used as path planning material.

[0028] Preferably, step S5 specifically includes the following process:

[0029] Subtract the original map matrix from the new map matrix to obtain the new obstacle coordinates, and extract the connected domain edge, centroid, and external frame formed by the new obstacle.

[0030] Preferably, the step S6 specifically includes the following process:

[0031] (6-1) Search for the original map road network node closest to the centroid of the new connected domain;

[0032] (6-2) Determine whether the node is within the new barrier connectivity domain outer frame. If so, proceed to step S7; if not, proceed to step (6-3);

[0033] (6-3) Determine whether the line connecting the node and its adjacent nodes intersects with the external frame. If so, proceed to step S7; if not, do nothing.

[0034] Preferably, the step S7 specifically includes the following process:

[0035] (7-1) Taking the centroid of the new obstacle as the center, the local map is gradually expanded by multiples of the length of the bounding box, and stops when three or more connected domains are detected in the local map;

[0036] (7-2) Eliminate the original road network nodes in the local map

[0037] (7-3) Find the communication nodes between the removed points and the original map by The removed points and exchange points are stored in the form of.

[0038] Preferably, the step S8 specifically includes the following process:

[0039] Repeat steps S3 and S4 for the local map to obtain local road network nodes.

[0040] Preferably, the step S9 specifically includes the following process:

[0041] (9-1) Traverse the exchange points obtained in step (7-3) and find the point in the local network node that is closest to the exchange point;

[0042] (9-2) Determine whether the line connecting the communication point in the original map and the corresponding nearest point in the local map intersects with the outer frame of the new connected domain. If so, the communication point does not participate in the fusion; if so, proceed to step (9-3);

[0043] (9-3) Eliminate the association between the original map communication point and the removed point, establish the association between the nearest point in the local map and the original map communication point, and use them as the new map adjacent nodes. Each of them stores the number, coordinates, and distance. The remaining original map nodes are not processed.

[0044] Compared with the prior art, the present invention has the following advantages:

[0045] Firstly, the present invention proposes a Voronoi diagram construction method based on line-surface generators, which uses the relationship between the shortest distance between adjacent obstacles and the obstacle bounding box to determine the discrete step length, follows the principle of densely discretizing nearby obstacles and sparsely discretizing distant obstacles, reduces the number of discrete points on the basis of fully discretizing the obstacle edges, and improves the efficiency of the algorithm.

[0046] Secondly, the present invention can adapt to the situation where new obstacles appear in the original map. In this case, there is no need to recalculate the global road network, but only generate a local road network with adaptive size for the newly added obstacles, so that the local road network can be seamlessly integrated with the global road network without destroying most of the global road network nodes, greatly improving the efficiency of the algorithm;

[0047] Thirdly, the algorithm proposed in the present invention is based on the characteristics of Delaunay triangulation, and the distance between the proposed road network and obstacles is appropriate, which is a safe road network;

[0048] Finally, the present invention provides the number and distance of each road network node and its adjacent nodes, which can quickly obtain the path index. Combined with the path planning algorithm, it can limit the path search space and improve the planning efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a flow chart of the incremental Voronoi network construction method of the present invention;

[0050] Figure 2 The original Voronoi network diagram generated in the embodiment of the present invention;

[0051] Figure 3 This is an incremental Voronoi road network diagram generated when a new obstacle appears in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The present invention will be further explained below in conjunction with the accompanying drawings and specific implementation methods. It should be understood that the following specific implementation methods are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

[0053] The incremental Voronoi network construction method described in the present invention implements the following process: Figure 1 As shown, the process mainly includes the following steps:

[0054] Step S1: number the obstacles based on the connected domains in the map, and obtain the edges, centroids and bounding boxes of the connected domains.

[0055] Preprocess the imported binary map: use functions to distinguish and number each connected domain, extract the edge, centroid and external frame coordinates of each connected domain, and concatenate the centroid coordinates into an array.

[0056] Step S2, triangulate the image using the centroid of the connected domain, and pair the regions according to the three-edge relationship.

[0057] The image is Delaunay triangulated using the centroid coordinate array. The triangle vertices correspond to the numbers of three adjacent connected domains. The connected domains are paired according to the triangle co-edge relationship. The connected domains at both ends of the same edge are a pair. In theory, a connected domain can be paired multiple times.

[0058] Step S3: Discretize the edges of each connected domain, construct a Voronoi graph based on the discrete edge point set, and obtain a road network node set consisting of Voronoi vertices.

[0059] (3-1) The edges of each connected domain are discretized, and the discrete step length is related to the distance between the paired elements, expressed as:

[0060]

[0061] Among them, d is the shortest distance between paired elements, L1 and L2 are the side length sets of paired elements, and λ is the discrete step length.

[0062] If the minimum distance between paired elements is less than or equal to the sum of the short sides in the external frames of the two elements, the inner elements use the minimum distance as the discrete steps; if the minimum distance between paired elements is greater than the sum of the short sides in the external frames of the two elements, the inner elements use twice the minimum distance as the discrete steps.

[0063] (3-2) Using the discretized edge point set as the generating point set, generate the point-based Voronoi diagram and Voronoi vertex coordinates;

[0064] (3-3) Merge the Voronoi regions generated by the same connected domain, that is, remove redundant vertices. Remove vertices outside the map range and vertices within the connected domain. If the line segments generated by two adjacent vertices cross the connected domain, that is, the line connecting the vertices intersects with the line connecting the adjacent discrete points of the connected domain, then remove the two vertices, and the remaining vertices constitute the road network nodes in the map, which are numbered.

[0065] Step S4, associate each road network node with its adjacent point according to the vertex common edge relationship to obtain the original map Voronoi road network.

[0066] Each road network node is associated with its neighboring points according to the common edge relationship of the original Voronoi vertices, and the neighboring point number, coordinates, and distance of each node are stored, which can be used as path planning material.

[0067] In this embodiment, the above steps are completed to generate the following effect: Figure 2 The Vino road network diagram is shown.

[0068] Step S5: Compare the new map with the original map to obtain the coordinates of the new obstacle, and extract the connected domain edge, centroid, and external frame of the new obstacle.

[0069] Subtract the original map matrix from the new map matrix to obtain the new obstacle coordinates, and extract the connected domain edge, centroid, and external frame formed by the new obstacle.

[0070] Step S6, determining whether the original map road network crosses the external frame formed by the new obstacle, if so, proceeding to step S7, if not, no processing is performed.

[0071] (6-1) Search for the original map road network node closest to the centroid of the new connected domain;

[0072] (6-2) Determine whether the node is within the new obstacle external frame. If so, proceed to step S7; if not, proceed to step (6-3);

[0073] (6-3) Determine whether the line connecting the node and its adjacent nodes intersects with the external frame. If so, proceed to step S7; if not, do nothing.

[0074] Step S7, taking the new obstacle centroid as the center, gradually expanding the local map by multiples of the length and width of the outer frame until there are three or more connected domains in the local map, and removing the original road network nodes in the local map.

[0075] (7-1) Taking the centroid of the new obstacle as the center, the local map is gradually expanded by multiples of the length of the bounding box, and stops when three or more connected domains are detected in the local map;

[0076] (7-2) Eliminate the original road network nodes in the local map

[0077] (7-3) Find the communication nodes between the removed points and the original map by The removed points and exchange points are stored in the form of.

[0078] Step S8, repeating steps S3 and S4 on the local map to obtain local road network nodes, and numbering the newly added nodes in the local map starting from the last digit of the original node number.

[0079] Step S9, merging the road network in the local map with the road network in the original map.

[0080] (9-1) Traverse the exchange points obtained in step (7-3) and find the point in the local network node that is closest to the exchange point;

[0081] (9-2) Determine whether the line connecting the communication point in the original map and the corresponding nearest point in the local map intersects with the outer frame of the new connected domain. If so, the communication point does not participate in the fusion; if so, proceed to step (9-3);

[0082] (9-3) Eliminate the association between the original map communication point and the removed point, establish the association between the nearest point in the local map and the original map communication point, and use them as the adjacent nodes of the new map. The relevant nodes store their own numbers, coordinates, and distances, and the remaining original map nodes are not processed.

[0083] In this embodiment, the above steps are completed to generate the following effect: Figure 3 The incremental Voronoi road network diagram is shown, where the interior of the box is divided into a local newly added road network.

[0084] The technical means disclosed in the scheme of the present invention are not limited to the technical means disclosed in the above-mentioned implementation mode, but also include technical schemes composed of any combination of the above technical features.

Claims

1. An incremental Voronoi road network construction method, characterized in that: The following steps are involved: S1: Number the obstacles based on the connected domains in the map, and obtain the edge, centroid and external frame of the connected domains; S2: triangulate the image using the centroid of the connected domain and pair the regions according to the three-edge relationship; S3: Discretize the edges of each connected domain, construct a Voronoi graph based on the discrete edge point set, and obtain a road network node set consisting of Voronoi vertices; S4: Associating each road network node with its adjacent point according to the vertex common edge relationship to obtain the original map Voronoi road network; S5: Compare the new map with the original map to obtain the coordinates of the new obstacle, and extract the edge, centroid, and external frame of the connected domain formed by the new obstacle; S6: Determine whether the original map road network crosses the external frame formed by the new obstacle. If so, proceed to step S7; if not, do not process; S7: Taking the centroid of the new obstacle as the center, the local map is gradually expanded by the multiple of the length and width of the external frame until there are three or more connected domains in the local map, and the original road network nodes in the local map are removed; S8: Repeat steps S3 and S4 for the local map to obtain local road network nodes; S9: Merge the road network in the local map with the road network in the original map.

2. The incremental Voronoi road network construction method according to claim 1, characterized in that: The step S1 specifically includes the following process: Preprocess the imported binary map, use functions to distinguish each connected domain, extract the edge, centroid and external frame coordinates of each connected domain, and concatenate the centroid coordinates into an array.

3. The incremental Voronoi road network construction method according to claim 1, characterized in that: The step S2 specifically includes the following process: The image is Delaunay triangulated using the centroid coordinate array. The triangle vertices correspond to the numbers of three adjacent connected domains. The connected domains are paired according to the triangle co-edge relationship, and the connected domains at both ends of the same edge are a pair.

4. The incremental Voronoi road network construction method according to claim 1, characterized in that: The step S3 specifically The process includes: (3-1) The edges of each connected domain are discretized, and the discrete step length is related to the distance between the paired elements, expressed as: Where d is the shortest distance between paired elements, L1 and L2 are the side length sets of paired elements, and λ is the discrete step length; (3-2) Taking the discrete edge point set as the generating point set, generate the Voronoi diagram and Voronoi vertex coordinates; (3-3) Merge the Voronoi regions generated by the same connected domain, that is, remove redundant vertices, including vertices outside the map and vertices within the connected region. If the line segment generated by two adjacent vertices crosses the connected region, the two vertices are removed, and the remaining vertices constitute the road network nodes in the map, which are numbered.

5. The incremental Voronoi road network construction method according to claim 1, characterized in that: The step S4 specifically includes the following process: Each road network node is associated with its neighboring points according to the common edge relationship of the original Voronoi vertices, and the neighboring point number, coordinates, and distance of each node are stored, which can be used as path planning material.

6. The incremental Voronoi road network construction method according to claim 1, characterized in that: The step S5 specifically includes the following process: subtracting the original map matrix from the new map matrix to obtain the coordinates of the new obstacle, and extracting the edge, centroid and external frame of the connected domain formed by the new obstacle.

7. The incremental Voronoi road network construction method according to claim 1, characterized in that: The step S6 specifically includes the following process: (6-1) Search for the original map road network node closest to the centroid of the new connected domain; (6-2) Determine whether the node is within the new barrier connectivity domain outer frame. If so, proceed to step S7; If not, proceed to step (6-3); (6-3) Determine whether the line connecting the node and its adjacent nodes intersects with the external frame. If so, proceed to step S7; if not, do nothing.

8. The incremental Voronoi road network construction method according to claim 1, characterized in that: The step S7 specifically The process includes: (7-1) Taking the centroid of the new obstacle as the center, the local map is gradually expanded by multiples of the length of the bounding box, and stops when three or more connected domains are detected in the local map; (7-2) Eliminate the original road network nodes in the local map (7-3) Find the communication nodes between the removed points and the original map by The removed points and exchange points are stored in the form of.

9. The incremental Voronoi road network construction method according to claim 1, characterized in that: The step S8 specifically The method includes the following steps: repeating steps S3 and S4 on the local map to obtain local road network nodes.

10. The incremental Voronoi road network construction method according to claim 1, characterized in that: The step S9 specifically The process includes: (9-1) Traverse the exchange points obtained in step (7-3) and find the point in the local network node that is closest to the exchange point; (9-2) Determine whether the line connecting the communication point in the original map and the corresponding nearest point in the local map intersects with the outer frame of the new connected domain. If so, the communication point does not participate in the fusion; if so, proceed to step (9-3); (9-3) Eliminate the association between the original map communication point and the removed point, establish the association between the nearest point in the local map and the original map communication point, and use them as the new map adjacent nodes. Each of them stores the number, coordinates, and distance. The remaining original map nodes are not processed.