A road network generation method and system, an electronic device and a storage medium

Abstract

The application provides a road network generation method and system, electronic equipment and a storage medium. The application constructs historical driving track data into track vectors, divides the track vectors into a plurality of grid track vector sets based on a world coordinate grid, constructs a single-pixel road network topology using a center line extraction technology, acquires Node and Link in the single-pixel road network topology, and completes construction of road network data, thereby realizing automatic processing of driving track data into map network data, and further realizing a map compilation function of a map road network without manual intervention, greatly improving the speed of map compilation and improving the timeliness of the road network.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, more particularly, to a road network generation method and system, an electronic device and a storage medium. BACKGROUND

[0002] Road network refers to a road system composed of various roads in a certain area, which are interconnected and interwoven into a network distribution. The node and link in the map represent this network road topology. Link refers to a section of road, which is the basic unit of the road model in the navigation system, and node is the end point of link or the intersection between links.

[0003] In the field of automatic driving of high-precision map services, there is a very high requirement for the real-time accuracy of the map. High-precision maps must be updated at a hierarchical or even second level in the future, and professional surveying and mapping teams are difficult to meet this new demand. The mode of "taking from the car and using for the car" has become the industry-recognized future through a large number of automatic driving vehicles running on the road to collect map data. This data collection method is low-cost, high-precision and large in quantity. Therefore, how to realize the automatic processing and extraction of map road network is a technical problem to be solved. SUMMARY

[0004] The present application provides a road network generation method, system, electronic device and storage medium to solve the problem of how to realize the automatic processing and extraction of map road network.

[0005] According to a first aspect of the present application, a road network generation method is provided, comprising:

[0006] Based on historical driving trajectory data, a trajectory vector set is constructed, and the trajectory vector set is corresponded to a preset world coordinate grid to obtain a trajectory vector set of a plurality of grids;

[0007] Based on the trajectory vector set of each grid, a center line extraction algorithm is used to obtain a single-pixel road network topology corresponding to each grid;

[0008] The pixel coordinates of each Node in the single-pixel road network topology are extracted to obtain a Node pixel coordinate set;

[0009] Based on the Node pixel coordinate set, the pixel point coordinates of the Link shape point string of all roads in the corresponding grid are extracted to obtain a Link pixel point coordinate set;

[0010] The Link pixel point coordinate set is summarized to obtain a Link shape point string set in the preset world coordinate grid;

[0011] In the preset world coordinate network, longitude and latitude coordinates of the set of Node pixel coordinates and the set of Link shape point strings are calculated and stored in the single-pixel road network topology generated road network data.

[0012] Based on the above technical solutions, the application can also be improved as follows.

[0013] Preferably, each trajectory vector in the set of trajectory vectors is composed of the start point coordinates, the end point coordinates, the ray angle from the start point coordinates to the end point coordinates, and the distance from the start point coordinates to the end point coordinates of each trajectory data in the historical driving trajectory data.

[0014] Preferably, the step of extracting the pixel coordinates of each Node in the single-pixel road network topology to obtain the set of Node pixel coordinates comprises:

[0015] The grid corresponding to the single-pixel road network topology is set as a grid map;

[0016] In the single-pixel road network topology, the number of adjacent points of each pixel point in the grid map is obtained;

[0017] When the number is equal to 1 or greater than 2, the corresponding pixel point is set as a Node, and the pixel coordinates of the Node are obtained;

[0018] The pixel coordinates of the Node in the single-pixel road network topology are summarized to obtain the set of Node pixel coordinates.

[0019] Preferably, the step of setting the corresponding pixel point as a Node and obtaining the pixel coordinates of the Node when the number is equal to 1 or greater than 2 comprises:

[0020] When the number is equal to 1, the corresponding pixel point is set as a road end point Node, and the pixel coordinates of the road end point Node are obtained;

[0021] When the number is greater than 2, the corresponding pixel point is set as a road intersection Node, and the pixel coordinates of the road intersection Node are obtained.

[0022] Preferably, the step of extracting the pixel coordinates of all roads in the corresponding grid based on the set of Node pixel coordinates to obtain the set of Link pixel coordinates comprises:

[0023] The grid corresponding to the single-pixel road network topology is set as a grid map;

[0024] In the single-pixel road network topology, the adjacent points of each pixel point of each topological branch are sequentially obtained;

[0025] when the adjacent point is in the Node pixel coordinate set, setting the adjacent point as a shape point of its corresponding topological branch;

[0026] collecting shape points of all topological branches in the single-pixel road network topology to obtain a Link shape point string and a corresponding Link pixel coordinate set of all roads.

[0027] Preferably, after the step of collecting the Link pixel coordinate set to obtain the Link shape point string set in the preset world coordinate grid, the method comprises:

[0028] optimizing the Link shape point string set based on a Douglas-Peucker algorithm.

[0029] Preferably, after the step of collecting the Link pixel coordinate set to obtain the Link shape point string set in the preset world coordinate grid, the method comprises:

[0030] storing a Link shape point string in the Link shape point string set in a grid where a first Node corresponding to the Link shape point string is located, and storing a corresponding number in a non-first Node corresponding to the Link shape point string when the Link shape point string spans multiple grids.

[0031] According to a second aspect of the present application, a road network generation system is provided, comprising:

[0032] a trajectory mapping module configured to construct a trajectory vector set based on historical driving trajectory data, and to correspond the trajectory vector set to a preset world coordinate grid to obtain a trajectory vector set of multiple grids;

[0033] a topology construction module configured to obtain a single-pixel road network topology corresponding to each grid using a centerline extraction algorithm based on the trajectory vector set of each grid;

[0034] a Node extraction module configured to extract a pixel coordinate of each Node in the single-pixel road network topology to obtain a Node pixel coordinate set;

[0035] a Link extraction module configured to extract a pixel coordinate of a Link shape point string of all roads in a grid corresponding to the Node pixel coordinate set to obtain a Link pixel coordinate set;

[0036] a coordinate collection module configured to collect the Link pixel coordinate set to obtain a Link shape point string set in the preset world coordinate grid;

[0037] a road network construction module, configured to calculate longitude and latitude coordinates of the Node pixel coordinate set and the Link point string set in the preset world coordinate network, and store the longitude and latitude coordinates into the single-pixel road network topology generated road network data.

[0038] According to a third aspect of the present application, an electronic device is provided, comprising a memory and a processor, the processor being configured to implement the steps of any of the road network generation methods of the first aspect when executing a computer management program stored in the memory.

[0039] According to a fourth aspect of the present application, a computer readable storage medium is provided, having a computer management program stored thereon, the computer management program being configured to implement the steps of any of the road network generation methods of the first aspect when executed by a processor.

[0040] The present application provides a road network generation method, system, electronic device and storage medium, the method comprising: constructing a trajectory vector set based on historical driving trajectory data, corresponding the trajectory vector set to a preset world coordinate grid to obtain a trajectory vector set of multiple grids; obtaining a single-pixel road network topology corresponding to each grid using a centerline extraction algorithm based on the trajectory vector set of each grid; extracting the pixel coordinates of each Node in the single-pixel road network topology to obtain a Node pixel coordinate set; extracting the pixel coordinates of the Link point string of all roads in the corresponding grid based on the Node pixel coordinate set to obtain a Link pixel coordinate set; summarizing the Link pixel coordinate set to obtain a Link point string set in the preset world coordinate grid; and calculating the longitude and latitude coordinates of the Node pixel coordinate set and the Link point string set in the preset world coordinate network, and storing the longitude and latitude coordinates into the single-pixel road network topology generated road network data. The present application constructs trajectory vectors from historical driving trajectory data, and divides the trajectory vectors into a trajectory vector set of multiple grids based on a world coordinate grid, constructs a single-pixel road network topology using a centerline extraction technique, and obtains Nodes and Links in the single-pixel road network topology to complete the construction of road network data, thereby realizing automatic processing of driving trajectory data into map network data, and further realizing the map compilation function of the map road network without manual intervention, greatly improving the speed of map compilation and improving the timeliness of the road network. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 A road network generation method flowchart is provided for the present application;

[0042] Figure 2 A diagram for calculating the angle of the Link connected to the Node is provided for the present application;

[0043] Figure 3 A schematic diagram of a method for extracting road endpoints and intersections on a grid map provided by the present application is shown in the following figure:

[0044] Figure 4 A schematic diagram of a Link-shaped point string across the grid provided by the present application is shown in the following figure:

[0045] Figure 5 A schematic diagram of the position adjustment of the boundary points of the Link-shaped point string provided by the present application is shown in the following figure:

[0046] Figure 6 A schematic diagram of the structure of a road network generation system provided by the present application is shown in the following figure:

[0047] Figure 7 A schematic diagram of the hardware structure of a possible electronic device provided by the present application is shown in the following figure:

[0048] Figure 8 A schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present application is shown in the following figure. DETAILED DESCRIPTION

[0049] The specific embodiments of the present application will be further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application, but are not used to limit the scope of the present application.

[0050] Figure 1 A flow chart of a road network generation method provided by the present application is shown in the following figure: Figure 1 As shown in the figure, the method comprises:

[0051] Step S100: constructing a trajectory vector set based on historical driving trajectory data, corresponding the trajectory vector set to a preset world coordinate grid, and obtaining trajectory vector sets of multiple grids;

[0052] It should be noted that the execution subject of the method of the present embodiment can be a computer terminal device with data processing, network communication and program running functions, such as a computer, a tablet computer, etc. It can also be a server device with the same or similar functions, and it can also be a cloud server with similar functions. The present embodiment does not limit this. In order to facilitate understanding, the present embodiment and the following embodiments will be described taking a server device as an example.

[0053] It can be understood that, in order to further improve the accuracy of the final road network data, the historical driving trajectory data used here does not include the driving trajectory data of vehicles on viaducts, tunnels and ramps.

[0054] It should be understood that the historical driving track data is processed into a set of track vectors, each of which is composed of two coordinates, the first coordinate of the vector is the starting point position, and the second is the end point position. The direction of the vector is represented by the angle of the ray from the first coordinate to the second coordinate. The length of the vector is the distance between the two coordinates.

[0055] In a specific implementation, the global (longitude from -180 degrees to +180 degrees, latitude from -90 degrees to +90 degrees) is evenly divided into (360 / M)*(180 / M) grids with M degrees as the grid width. The track vectors are distributed into the corresponding grids according to the grid range, and the vectors that cross the grids are divided into two vectors with the crossing grid point as the division point, and finally a set of track vectors belonging to different grids is formed. Wherein M degrees can be set according to specific needs, and the embodiment does not limit this.

[0056] Step S200: based on the set of track vectors of each grid, using a center line extraction algorithm to obtain the single-pixel road network topology corresponding to each grid;

[0057] In a specific implementation, all grids are traversed in parallel, and a grid map is created for each grid (where a single pixel occupies a longitude and latitude value of R). And draw the track vectors in the grid map (the image coordinates are equal to the track vector coordinates minus the grid lower left corner coordinates and then divided by R). Then use the center line extraction algorithm to converge the roads with width drawn on the grid map into a road network topology with only 1 pixel width.

[0058] Step S300: extracting the pixel coordinates of each Node in the single-pixel road network topology to obtain a set of Node pixel coordinates;

[0059] In a specific implementation, the grid map of all grids is traversed in parallel, and the pixel coordinates of the Node are extracted. The pixel coordinates of the Node are multiplied by R and then added to the longitude and latitude coordinates of the grid lower left point to obtain their longitude and latitude coordinates. Then according to the coordinate position of the Node in the grid, it is numbered from top to bottom and from left to right (in-grid numbering). The number of each Node (numbered in the map) is composed of the grid number of the Node and its in-grid number.

[0060] Step S400: based on the set of Node pixel coordinates, extracting the pixel coordinates of the Link shape point string of all roads in the corresponding grid to obtain a set of Link pixel coordinates;

[0061] In a specific implementation, all grid raster maps are traversed in parallel to extract the pixel coordinates of the Link-shaped points. The pixel coordinates of the Link-shaped points are multiplied by R and then added to the longitude and latitude coordinates of the lower-left point of the grid to obtain the longitude and latitude coordinates of the Link-shaped points. A circle is drawn with the Node as the center and L as the radius, the circle intersects the Link connected to the Node at point W, and the angle of the vector from the Node to W (Link angle) is calculated. All Links connected to the Node are numbered in ascending order of Link angle to obtain the Link angle number of the Link at the Node. For specific effects, refer to Figure 2 The number of each Link (which is unique in the map) is composed of the number of the first Node of the Link and the angle number of the Link at the first Node.

[0062] Step S500: The Link pixel point coordinate set is summarized to obtain a Link-shaped point string set in the preset world coordinate grid.

[0063] In a specific implementation, all grids are traversed in parallel, all Link-shaped point strings in the road network are summarized, and the Douglas-Peucker algorithm is used for shape point compression and smoothing to obtain optimized shape point strings. Finally, the shape point strings of the Link are saved in the grid where the first Node is located; for the Link that spans multiple grids, the number of the Link is saved in the grid where the non-first Node is located.

[0064] Step S600: In the preset world coordinate network, the longitude and latitude coordinates of the Node pixel coordinate set and the Link-shaped point string set are calculated and stored in the single-pixel road network topology generation road network data.

[0065] It can be understood that based on the defects in the background art, the embodiment of the present application proposes a road network generation method. The method comprises: constructing a trajectory vector set based on historical driving trajectory data, corresponding the trajectory vector set to a preset world coordinate grid to obtain a trajectory vector set of multiple grids; based on the trajectory vector set of each grid, using a center line extraction algorithm to obtain a single-pixel road network topology structure corresponding to each grid; extracting the pixel coordinates of each Node in the single-pixel road network topology structure to obtain a Node pixel coordinate set; based on the Node pixel coordinate set, extracting the pixel point coordinates of the Link shape point string of all roads in the corresponding grid to obtain a Link pixel point coordinate set; summarizing the Link pixel point coordinate set to obtain a Link shape point string set in the preset world coordinate grid; in the preset world coordinate network, calculating the latitude and longitude coordinates of the Node pixel coordinate set and the Link shape point string set, and storing them to the single-pixel road network topology structure to generate road network data. The present application constructs historical driving trajectory data into trajectory vectors and divides them into multiple grid trajectory vector sets based on a world coordinate grid, uses center line extraction technology to construct a single-pixel road network topology structure, and obtains the Node and Link in the single-pixel road network topology structure to complete the construction of road network data, thereby realizing automatic processing of driving trajectory data to map network data, and further realizing the map compilation function of the map road network without manual intervention, greatly improving the speed of map compilation and improving the timeliness of the road network.

[0066] In a possible embodiment, further, the step of extracting the Node pixel coordinates in the single-pixel road network topology structure comprises:

[0067] Step S301: setting the grid corresponding to the single-pixel road network topology structure as a grid map;

[0068] Step S302: in the single-pixel road network topology structure, obtaining the number of adjacent points of each pixel point in the grid map;

[0069] Step S303: when the number is equal to 1 or greater than 2, setting the corresponding pixel point as a Node and obtaining the pixel coordinates of the Node;

[0070] Further, when the number is equal to 1, setting the corresponding pixel point as a road end point Node and obtaining the pixel coordinates of the road end point Node; when the number is greater than 2, setting the corresponding pixel point as a road intersection Node and obtaining the pixel coordinates of the road intersection Node.

[0071] Step S304: aggregating the pixel coordinates of the Node in the single-pixel road network topology to obtain a Node pixel coordinate set.

[0072] Referring to Figure 3 , Figure 3 A schematic diagram of a method for extracting road endpoints and intersection points on a grid map provided by the present application; wherein the specific method for extracting road endpoints and intersection points can be that, in the grid map with a road network, the width of the grid map is set as W and the height is set as H; the grid map with a road network is black and the grid map without a road network is white.

[0073] In order to further improve the accuracy of Node extraction, it is also necessary to process the boundary points in the grid map, and sequentially take out each pixel point in the single-pixel road network topology, initialize the coordinates (X, Y) of the pixel point as (0, 0), initialize the pixel coordinate list A, and obtain the eight adjacent points of the pixel point, wherein if the pixel point is a boundary point of the grid map, the eight adjacent points thereof are processed as follows:

[0074] When the pixel point is an upper boundary point in the grid map, if Y is greater than 0, then (X, Y-1) is put into the list A;

[0075] When the pixel point is a right upper boundary vertex in the grid map, if X is greater than 0, then (X-1, Y-1) is put into the list A;

[0076] When the pixel point is a left upper boundary vertex in the grid map, if X is less than W-1, then (X+1, Y-1) is put into the list A;

[0077] When the pixel point is a right boundary point in the grid map, if X is greater than 0, then (X-1, Y) is put into the list A;

[0078] When the pixel point is a left boundary point in the grid map, if X is less than W-1, then (X+1, Y) is put into the list A;

[0079] When the pixel point is a lower boundary point in the grid map, if Y is less than H-1, then (X, Y+1) is put into the list A;

[0080] When the pixel point is a right lower boundary vertex in the grid map, if X is greater than 0, then (X-1, Y+1) is put into the list A;

[0081] When the pixel point is a left lower boundary vertex in the grid map, if X is less than W-1, then (X+1, Y+1) is put into the list A.

[0082] After the above boundary point processing, the number of pixels with black color in the list A and N, which is the adjacent point corresponding to the pixel coordinate (X, Y), are counted; if N is equal to 1, the point (X, Y) is a road end point Node, and if N>2, the point (X, Y) is a cross point Node. Each pixel point in the single-pixel road network topology structure is taken out in turn until all the pixel points are judged.

[0083] In the embodiment, the cross points and end points in the single-pixel road network topology structure are extracted in the grid map, thereby realizing automatic extraction of the Node in the driving track, greatly improving the speed of map compilation, and improving the timeliness of the road network.

[0084] In a possible implementation manner, the step of extracting, based on the Node pixel coordinate set, pixel point coordinates of all road Link shape point strings in the corresponding grid, to obtain a Link pixel point coordinate set, includes:

[0085] Step S401: set the grid corresponding to the single-pixel road network topology structure as a grid map;

[0086] Step S401: in the single-pixel road network topology structure, the adjacent points of each pixel point of each topology branch are obtained in turn;

[0087] Step S401: when the adjacent point is in the Node pixel coordinate set, the adjacent point is set as a shape point of the corresponding topology branch;

[0088] Step S401: the shape points of all topology branches in the single-pixel road network topology structure are summarized to obtain Link shape point strings of all roads and a corresponding Link pixel point coordinate set.

[0089] In a specific implementation, in the grid map with the road network, the width of the grid map is set as W and the height is set as H; the grid map with the road network is black and the grid map without the road network is white; the Node pixel coordinate set in the single-pixel road network topology interface is T.

[0090] Initialize the result set F, the intermediate list S and the pixel coordinate list A, take a pixel point P from the set T, insert P into the tail of the list S, and set the coordinates (X, Y) of the initial pixel point P as (0, 0); obtain the eight adjacent points of the pixel point P, wherein if the pixel point is a boundary point of the grid map, the eight adjacent points are processed as follows:

[0091] When the pixel point is an upper boundary point in the grid map, if Y is greater than 0, (X, Y-1) is put into the list A;

[0092] When the pixel point is a top-right boundary vertex in the grid map, if X is greater than 0, then (X-1, Y-1) is put into list A;

[0093] When the pixel point is a top-left boundary vertex in the grid map, if X is less than W-1, then (X+1, Y-1) is put into list A;

[0094] When the pixel point is a right boundary point in the grid map, if X is greater than 0, then (X-1, Y) is put into list A;

[0095] When the pixel point is a left boundary point in the grid map, if X is less than W-1, then (X+1, Y) is put into list A;

[0096] When the pixel point is a bottom boundary point in the grid map, if Y is less than H-1, then (X, Y+1) is put into list A;

[0097] When the pixel point is a bottom-right boundary vertex in the grid map, if X is greater than 0, then (X-1, Y+1) is put into list A;

[0098] When the pixel point is a bottom-left boundary vertex in the grid map, if X is less than W-1, then (X+1, Y+1) is put into list A.

[0099] After the boundary point processing, another pixel point with a black color is searched in list A, if the pixel point is in set T, the pixel point is inserted into the end of list S, the coordinates of the pixel point P are replaced by the coordinates of the pixel point B, and the pixel point B is set to white, and set T is traversed until all pixel points in set T are traversed, and S is inserted into set F, to obtain a set of Link-shaped point strings in the road network in the grid map.

[0100] In the embodiment, the Link-shaped point strings in the single-pixel road network topology are extracted in the grid map, so that the automatic extraction of the Link-shaped point strings in the driving track is realized, the speed of map compilation is greatly improved, and the timeliness of the road network is improved.

[0101] In a possible implementation manner, the step of aggregating the Link pixel point coordinate set to obtain the Link-shaped point string set in the preset world coordinate grid further includes:

[0102] A Link-shaped point string with an end point at a grid boundary in the Link pixel point coordinate set is obtained;

[0103] Based on the latitude and longitude value R of the single pixel point, a Link-shaped point string adjacent to the end point is calculated and set as a Link-shaped point string in the preset world coordinate grid;

[0104] In a specific implementation, referring to Figure 4 , Figure 4 The schematic diagram of the Link-shaped point string across the grid provided by the present application; the longitude and latitude value of a single pixel point is R, when the end point in the Link row point string A is on the grid boundary, the end point coordinates are added with R, so as to calculate the adjacent Link row point string B of the end point, and the Link row point string B and the Link row point string A are set as a Link-shaped point string in the preset world coordinate grid.

[0105] Based on the longitude and latitude value R of a single pixel point, the end point of the Link-shaped point string without adjacent end point is corrected.

[0106] In a specific implementation, referring to Figure 5 , Figure 5 The schematic diagram of the Link-shaped point string boundary point position adjustment provided by the present application, wherein the circle represents the Link head point, the square represents the middle-shaped point, the pentagon represents the Link tail point, and the triangle represents the boundary Node; the step of correcting the end point can be that, when judging that the end point P (the head point and the tail point) of the Link is on the grid boundary, the longitude and latitude value R of a single pixel point is obtained, if the X coordinate of the boundary point P is on the left edge of the grid, X=X+R; if the X coordinate of the boundary point P is on the right edge of the grid, X=X-R; if the Y coordinate of the boundary point P is on the lower edge of the grid, Y=Y+R; if the Y coordinate of the boundary point P is on the lower edge of the grid, Y=Y-R.

[0107] In the embodiment, the Link-shaped point string is found in the gridded road network map, so that the automatic collection of the Link-shaped point string is realized, the speed of map compilation is greatly improved, and the timeliness of the road network is improved.

[0108] Please refer to Figure 6 , Figure 6 The structural diagram of a road network generation system provided by the embodiment of the present application, as shown in Figure 6 The road network generation system comprises a trajectory mapping module 100, a topology construction module 200, a Node extraction module 300, a Link extraction module 400, a coordinate collection module 500 and a road network construction module 600, wherein:

[0109] The trajectory mapping module 100 is used for constructing a trajectory vector set based on historical driving trajectory data, corresponding the trajectory vector set to a preset world coordinate grid, and obtaining a trajectory vector set of a plurality of grids;

[0110] The topology construction module 200 is used for obtaining a single-pixel road network topology structure corresponding to each grid based on the trajectory vector set of each grid by using a center line extraction algorithm;

[0111] The Node extraction module 300 is used to extract the pixel coordinates of each Node in the single-pixel road network topology to obtain a set of Node pixel coordinates.

[0112] The Link extraction module 400 is used to extract the pixel coordinates of all the Link-shaped point strings of all roads in the corresponding grid based on the Node pixel coordinate set, and obtain the Link pixel coordinate set.

[0113] The coordinate summarization module 500 is used to summarize the set of Link pixel coordinates to obtain the set of Link-shaped point strings in the preset world coordinate grid.

[0114] The road network construction module 600 is used to calculate the latitude and longitude coordinates of the Node pixel coordinate set and the Link-shaped point string set in the preset world coordinate network, and store them in the single-pixel road network topology structure to generate road network data.

[0115] It is understood that the road network generation system provided by the present invention corresponds to the road network generation method provided in the foregoing embodiments. The relevant technical features of the road network generation system can be referred to the relevant technical features of the road network generation method, and will not be repeated here.

[0116] Please see Figure 7 , Figure 7 This is a schematic diagram illustrating an embodiment of the electronic device provided in this invention. For example... Figure 7 As shown, this embodiment of the invention provides an electronic device, including a memory 1310, a processor 1320, and a computer program 1311 stored in the memory 1310 and executable on the processor 1320. When the processor 1320 executes the computer program 1311, it performs the following steps:

[0117] A trajectory vector set is constructed based on historical driving trajectory data. This trajectory vector set is mapped to a preset world coordinate grid, resulting in multiple grid trajectory vector sets. Based on the trajectory vector set of each grid, a centerline extraction algorithm is used to obtain the single-pixel road network topology structure corresponding to each grid. The pixel coordinates of each Node in the single-pixel road network topology structure are extracted to obtain a Node pixel coordinate set. Based on the Node pixel coordinate set, the pixel coordinates of all Link-shaped point strings of roads in their corresponding grids are extracted to obtain a Link pixel coordinate set. The Link pixel coordinate sets are summarized to obtain a Link-shaped point string set in the preset world coordinate grid. In the preset world coordinate network, the latitude and longitude coordinates of the Node pixel coordinate set and the Link-shaped point string set are calculated and stored in the single-pixel road network topology structure to generate road network data.

[0118] Please refer to Figure 8 , Figure 8 An embodiment of a computer readable storage medium provided by the present application is shown. As shown in Figure 8 , the embodiment provides a computer readable storage medium 1400, which stores a computer program 1411, and the computer program 1411 is executed by a processor to implement the following steps:

[0119] Based on historical driving trajectory data, a trajectory vector set is constructed, the trajectory vector set is corresponded to a preset world coordinate grid, and a trajectory vector set of multiple grids is obtained; based on the trajectory vector set of each grid, a single-pixel road network topology corresponding to each grid is obtained using a center line extraction algorithm; pixel coordinates of each Node in the single-pixel road network topology are extracted, and a Node pixel coordinate set is obtained; based on the Node pixel coordinate set, pixel point coordinates of a Link shape point string of all roads in the corresponding grid are extracted, and a Link pixel point coordinate set is obtained; the Link pixel point coordinate set is summarized, and a Link shape point string set in the preset world coordinate grid is obtained; in the preset world coordinate network, latitude and longitude coordinates of the Node pixel coordinate set and the Link shape point string set are calculated, and are stored in the single-pixel road network topology generation road network data.

[0120] The application provides a road network generation method and system, an electronic device and a storage medium. The method comprises the following steps: constructing a trajectory vector set based on historical driving trajectory data, corresponding the trajectory vector set to a preset world coordinate grid to obtain a trajectory vector set of a plurality of grids, obtaining a single-pixel road network topology structure corresponding to each grid by using a center line extraction algorithm based on the trajectory vector set of each grid, extracting pixel coordinates of each Node in the single-pixel road network topology structure to obtain a Node pixel coordinate set, extracting pixel point coordinates of a Link shape point string of all roads in the corresponding grid based on the Node pixel coordinate set to obtain a Link pixel point coordinate set, summarizing the Link pixel point coordinate set to obtain a Link shape point string set in the preset world coordinate grid, calculating the latitude and longitude coordinates of the Node pixel coordinate set and the Link shape point string set in the preset world coordinate network, and storing the latitude and longitude coordinates in the single-pixel road network topology structure generation road network data. The application constructs trajectory vectors from historical driving trajectory data, divides the trajectory vectors into a plurality of trajectory vector sets of grids based on a world coordinate grid, constructs a single-pixel road network topology structure by using a center line extraction technology, and obtains Nodes and Links in the single-pixel road network topology structure to complete the construction of road network data, thereby realizing automatic processing of driving trajectory data into map network data, realizing map compilation of the map network without manual intervention, greatly improving the speed of map compilation, and improving the timeliness of the road network.

[0121] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0122] Those skilled in the art should understand that the embodiments of the application can be provided as a method, a system, or a computer program product. Therefore, the application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the application can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0123] The present application is described in reference to the flowchart illustrations and / or block diagrams according to the methods, apparatus (systems) and computer program products of embodiments of the application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.

[0124] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.

[0125] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.

[0126] While the preferred embodiments of the application have been described, additional variations and modifications can be made to the embodiments by those of skill in the art once they have the benefit of the present disclosure without departing from the spirit and scope of the application. Accordingly, the attached claims are intended to cover all such variations and modifications as falling within the scope of the application.

[0127] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A method of generating a road network, characterized by, The method comprises: Based on historical driving trajectory data, a trajectory vector set is constructed, the trajectory vector set is corresponded to a preset world coordinate grid, and a trajectory vector set of multiple grids is obtained; Based on the trajectory vector set of each grid, a center line extraction algorithm is used to obtain a single-pixel road network topology corresponding to each grid; Pixel coordinates of each Node in the single-pixel road network topology are extracted to obtain a Node pixel coordinate set; Based on the Node pixel coordinate set, pixel point coordinates of a Link shape point string of all roads in the corresponding grid are extracted to obtain a Link pixel point coordinate set; The Link pixel point coordinate set is summarized to obtain a Link shape point string set in the preset world coordinate grid; In the preset world coordinate grid, the latitude and longitude coordinates of the Node pixel coordinate set and the Link shape point string set are calculated, and are stored in the single-pixel road network topology to generate road network data.

2. The road network generation method according to claim 1, characterized by, Each trajectory vector in the trajectory vector set is composed of a start point coordinate, an end point coordinate, a ray angle from the start point coordinate to the end point coordinate, and a distance from the start point coordinate to the end point coordinate in the historical driving trajectory data.

3. The road network generation method according to claim 1, characterized by, The step of extracting pixel coordinates of each Node in the single-pixel road network topology to obtain a Node pixel coordinate set comprises: The grid corresponding to the single-pixel road network topology is set as a grid graph; In the single-pixel road network topology, the number of adjacent points of each pixel point in the grid graph is obtained; When the number is equal to 1 or greater than 2, the corresponding pixel point is set as a Node, and the pixel coordinates of the Node are obtained; The pixel coordinates of the Node in the single-pixel road network topology are summarized to obtain a Node pixel coordinate set.

4. The road network generation method according to claim 3, characterized by, The step of setting the corresponding pixel point as a Node and obtaining the pixel coordinates of the Node when the number is equal to 1 or greater than 2 comprises: When the number is equal to 1, the corresponding pixel point is set as a road end point Node, and the pixel coordinates of the road end point Node are obtained; When the number is greater than 2, the corresponding pixel point is set as a road intersection Node, and the pixel coordinates of the road intersection Node are obtained.

5. The road network generation method according to claim 1, characterized by, The step of extracting pixel point coordinates of a Link shape point string of all roads in the corresponding grid based on the Node pixel coordinate set to obtain a Link pixel point coordinate set comprises: The grid corresponding to the single-pixel road network topology is set as a grid graph; In the single-pixel road network topology, adjacent points of each pixel point of each topological branch are obtained in turn; When the adjacent point is in the Node pixel coordinate set, the adjacent point is set as a shape point of the corresponding topological branch; The shape points of all topological branches in the single-pixel road network topology are summarized to obtain a Link shape point string of all roads and a corresponding Link pixel point coordinate set.

6. The road network generation method according to claim 1, characterized by, After the step of aggregating the Link pixel point coordinate set to obtain the Link-shaped point string set in the preset world coordinate grid, the method comprises: optimizing the Link-shaped point string set based on a Douglas-Pok algorithm.

7. The road network generation method according to claim 1, characterized by, After the step of aggregating the Link pixel point coordinate set to obtain the Link-shaped point string set in the preset world coordinate grid, the method comprises: saving the Link-shaped point string in the Link-shaped point string set to the grid where the first Node corresponding to the Link-shaped point string is located, and saving the corresponding number in the non-first Node corresponding to the Link-shaped point string when the Link-shaped point string spans multiple grids.

8. A road network generation system characterized by comprising: The method comprises: a trajectory mapping module, configured to construct a trajectory vector set based on historical driving trajectory data, and correspond the trajectory vector set to a preset world coordinate grid to obtain trajectory vector sets of multiple grids; a topology construction module, configured to obtain a single-pixel road network topology structure corresponding to each grid based on the trajectory vector set of each grid using a center line extraction algorithm; a Node extraction module, configured to extract pixel coordinates of each Node in the single-pixel road network topology structure to obtain a Node pixel coordinate set; a Link extraction module, configured to extract pixel point coordinates of Link-shaped point strings of all roads in the grid corresponding to the Node pixel coordinate set to obtain a Link pixel point coordinate set; a coordinate aggregation module, configured to aggregate the Link pixel point coordinate set to obtain a Link-shaped point string set in the preset world coordinate grid; a road network construction module, configured to calculate latitude and longitude coordinates of the Node pixel coordinate set and the Link-shaped point string set in the preset world coordinate grid, and store the latitude and longitude coordinates to generate road network data of the single-pixel road network topology structure.

9. An electronic device, comprising: The method comprises a memory and a processor, and the processor is configured to execute a computer management program stored in the memory to implement the steps of the road network generation method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer management program is stored on the memory and is executed by the processor to implement the steps of the road network generation method according to any one of claims 1-7.

Images