A road network skeleton generation method and device, electronic equipment and storage medium

Abstract

The application relates to the technical field of automatic driving, in particular to a road network skeleton generation method and device, electronic equipment and a storage medium, which are used to improve the accuracy of road network skeleton generation. The method comprises the following steps: obtaining each trajectory data corresponding to a target map area; dividing each trajectory data obtained into each trajectory group, and respectively converting each trajectory group into a corresponding node in a directed graph; determining the connection relationship between each trajectory group based on the driving information of each trajectory data contained in each trajectory group, and converting the connection relationship between two connected trajectory groups into a directed edge between the corresponding two nodes in the directed graph; and obtaining the intersection skeleton, the road skeleton and the road network topology relationship corresponding to the target map area based on each node in the directed graph and the directed edges between the nodes. In this way, the directed graph is established based on massive trajectory data, the road network skeleton is generated based on the directed graph, and the accuracy of the road network skeleton is improved.

Description

TECHNICAL FIELD

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

[0002] High-precision map, also known as high-precision map, is one of the core technologies in the field of automatic driving. At present, the development of many functions of intelligent vehicle automatic driving is based on high-precision map, and the development of high-precision map directly affects the safety and accuracy of automatic driving. Compared with navigation electronic map, high-precision map can provide lane-level road conditions, with centimeter-level precision, higher update frequency of elements, and more detailed data model, which can help autonomous vehicles better avoid potential risks. At present, road data is usually fused based on a road network skeleton to generate a high-precision map, and the establishment of a road network skeleton is an important part of generating a high-precision map.

[0003] In the prior art, when establishing a road network skeleton, a manual labeling method is usually used to manually label on a point cloud map or a semantic map to establish a road network skeleton.

[0004] However, when generating a road network skeleton by the method in the related art, since manual labeling needs to rely on the professional quality of professionals, different professionals label different road nodes, resulting in inaccurate road network skeletons. For example, when the labeling experience of professionals is not enough, there may be labeling errors.

[0005] Therefore, in the related art, the accuracy of road network skeleton generation needs to be further improved. SUMMARY

[0006] Embodiments of the present application provide a road network skeleton generation method and device, an electronic device, and a storage medium to improve the accuracy of road network skeleton generation.

[0007] The specific technical solutions provided by the embodiments of the present application are as follows:

[0008] In a first aspect, a road network skeleton generation method is provided, comprising:

[0009] obtaining each trajectory data corresponding to a target map area, wherein each trajectory data contains a driving trajectory of a corresponding vehicle in the target map area;

[0010] dividing each trajectory data obtained into each trajectory group, and respectively converting each trajectory group into a corresponding node in a directed graph; wherein the similarity between each trajectory data contained in each trajectory group reaches a similarity threshold;

[0011] determine a connection relationship between the trajectory groups based on driving information of the trajectory data included in the trajectory groups, and convert the connection relationship between two connected trajectory groups into a directed edge between corresponding two nodes in a directed graph;

[0012] obtain a road skeleton, a road skeleton, and a road network topology relationship corresponding to the target map area based on the nodes in the directed graph and the directed edges between the nodes.

[0013] Optionally, the obtained trajectory data is divided into trajectory groups, including:

[0014] According to a preset region segmentation manner, the target map area is segmented to obtain a sub-region grid set corresponding to the target map area;

[0015] From the sub-region grid set, each trajectory grid is selected, wherein each trajectory grid includes at least one trajectory data in a trajectory data;

[0016] For each trajectory grid, the following operations are performed respectively: for a trajectory grid, the trajectory data in the trajectory data is clustered when the similarity reaches a first similarity threshold, and at least one target cluster is obtained.

[0017] Each target cluster obtained is taken as a corresponding trajectory group.

[0018] Optionally, the driving information of each trajectory data at least includes: a driving time corresponding to the corresponding trajectory data;

[0019] Then, based on the driving information of the trajectory data included in the trajectory groups, the connection relationship between the trajectory groups is determined, including:

[0020] For each trajectory group, the following operations are performed respectively:

[0021] If the first trajectory data included in a trajectory group is continuous with the second trajectory data included in another trajectory group, and the driving time of the second trajectory data is before the driving time of the first trajectory data, it is determined that the trajectory group corresponding to the second trajectory data is the predecessor connection of the trajectory group corresponding to the first trajectory data.

[0022] If the first trajectory data included in a trajectory group is continuous with the second trajectory data included in another trajectory group, and the driving time of the second trajectory data is after the driving time of the first trajectory data, it is determined that the trajectory group corresponding to the second trajectory data is the successor connection of the trajectory group corresponding to the first trajectory data.

[0023] Optionally, the connection relationship between two connected trajectory groups is converted into a directed edge between corresponding two nodes in a directed graph, including:

[0024] For each trajectory group, the following operations are performed:

[0025] one trajectory group is taken as a first trajectory group, one trajectory group which is successively connected to the first trajectory group is taken as a second trajectory group, and a connection relationship between the first trajectory group and the second trajectory group is converted into a directed edge between two corresponding nodes in the directed graph, wherein a direction of the directed edge is from the node corresponding to the second trajectory group to the node corresponding to the first trajectory group;

[0026] one trajectory group is taken as a first trajectory group, one trajectory group which is successively connected to the first trajectory group is taken as a third trajectory group, and a connection relationship between the first trajectory group and the third trajectory group is converted into a directed edge between two corresponding nodes in the directed graph, wherein a direction of the directed edge is from the node corresponding to the first trajectory group to the node corresponding to the third trajectory group.

[0027] Optionally, after the connection relationship between the two connected trajectory groups is converted into the directed edge between the two corresponding nodes in the directed graph, the method further comprises:

[0028] based on the similarity between each trajectory group, each node in the directed graph is merged to obtain new nodes;

[0029] For each node currently contained in the directed graph, the following operations are performed: if an out-degree of a node is greater than a first out-degree threshold, or an in-degree of a node is greater than a first in-degree threshold, a node is determined as a first node, wherein the out-degree of a node is a number of directed edges with the node as a starting point, and the in-degree of a node is a number of directed edges with the node as a terminal point;

[0030] each first node which is connected in the obtained first node set is merged into a corresponding second node.

[0031] Optionally, based on the similarity between each trajectory group, each node in the directed graph is merged to obtain new nodes, comprising:

[0032] For each trajectory grid, the following operations are performed: each trajectory group which has a similarity reaching a second similarity threshold and is not connected in a trajectory grid and eight neighborhood trajectory grids of the trajectory grid is clustered to obtain at least one trajectory group cluster;

[0033] each node corresponding to each trajectory group belonging to a same trajectory group cluster in the directed graph is merged into a corresponding new node.

[0034] Optionally, based on each node in the directed graph and the directed edges between the nodes, a road intersection skeleton, a road skeleton and a road network topology relationship corresponding to a target map region are obtained, comprising:

[0035] determine the road node and the intersection node in the directed graph based on the out-degree and the in-degree of each node in the directed graph;

[0036] generate the intersection skeleton corresponding to the target map area based on the trajectory group corresponding to each intersection node;

[0037] generate the road skeleton corresponding to the target map area based on the trajectory group corresponding to each road node;

[0038] generate the road network topology relationship corresponding to the target map area based on the connection relationship corresponding to the directed edge between each node.

[0039] Optionally, the determining the road node and the intersection node in the directed graph based on the out-degree and the in-degree of each node in the directed graph comprises:

[0040] The following operations are performed for each node in the directed graph:

[0041] If the out-degree of a node is greater than a second out-degree threshold, or the in-degree of a node is greater than a second in-degree threshold, the node is determined to be an intersection node.

[0042] If the out-degree of a node is not greater than the second out-degree threshold, and the in-degree of the node is not greater than the second in-degree threshold, the node is determined to be a road node.

[0043] In a second aspect, a road network skeleton generation apparatus is provided, comprising:

[0044] an acquisition module configured to acquire each trajectory data corresponding to a target map area, wherein each trajectory data comprises a driving trajectory of a corresponding vehicle in the target map area;

[0045] a grouping module configured to divide each trajectory data obtained into a trajectory group, and convert each trajectory group into a corresponding node in a directed graph; wherein the similarity between each trajectory data included in each trajectory group reaches a similarity threshold;

[0046] a first processing module configured to determine a connection relationship between each trajectory group based on driving information of each trajectory data included in each trajectory group, and convert the connection relationship between two connected trajectory groups into a directed edge between two corresponding nodes in the directed graph;

[0047] a generation module configured to obtain an intersection skeleton, a road skeleton and a road network topology relationship corresponding to the target map area based on each node in the directed graph and the directed edge between each node.

[0048] Optionally, when each trajectory data obtained is divided into a trajectory group, the grouping module is further configured to:

[0049] segment the target map region according to the preset region segmentation manner, to obtain a sub-region grid set corresponding to the target map region;

[0050] filter each trajectory grid from the sub-region grid set, wherein each trajectory grid contains a piece of trajectory data in at least one trajectory data;

[0051] For each trajectory grid, the following operations are performed respectively: clustering each piece of trajectory data in a trajectory grid whose similarity reaches a first similarity threshold, to obtain at least one target cluster;

[0052] respectively taking each obtained target cluster as a corresponding trajectory group.

[0053] Optionally, the driving information of each piece of trajectory data at least includes a driving moment corresponding to the corresponding trajectory data;

[0054] Then, when determining the connection relationship between the trajectory groups based on the driving information of the trajectory data contained in each trajectory group, the first processing module is further configured to:

[0055] For each trajectory group, the following operations are performed respectively:

[0056] If the first trajectory data contained in one trajectory group is continuous with the second trajectory data contained in another trajectory group, and the driving moment corresponding to the second trajectory data is before the driving moment corresponding to the first trajectory data, it is determined that the trajectory group corresponding to the second trajectory data is the predecessor connection of the trajectory group corresponding to the first trajectory data.

[0057] If the first trajectory data contained in one trajectory group is continuous with the second trajectory data contained in another trajectory group, and the driving moment of the second trajectory data is after the driving moment of the first trajectory data, it is determined that the trajectory group corresponding to the second trajectory data is the successor connection of the trajectory group corresponding to the first trajectory data.

[0058] Optionally, when converting the connection relationship between the two connected trajectory groups into a directed edge between the corresponding two nodes in the directed graph, the first processing module is further configured to:

[0059] For each trajectory group, the following operations are performed respectively:

[0060] taking one trajectory group as a first trajectory group, taking a trajectory group which is the predecessor connection of the first trajectory group as a second trajectory group, and converting the connection relationship between the first trajectory group and the second trajectory group into a directed edge between the corresponding two nodes in the directed graph, wherein the direction of the directed edge is from the node corresponding to the second trajectory group to the node corresponding to the first trajectory group;

[0061] transforming a connection relationship between the first track group and the third track group into a directed edge between two nodes corresponding to the first track group and the third track group in the directed graph, wherein a direction of the directed edge is from the node corresponding to the first track group to the node corresponding to the third track group.

[0062] Optionally, after transforming the connection relationship between the two connected track groups into the directed edge between the two nodes corresponding to the two track groups in the directed graph, the apparatus further comprises a second processing module configured to:

[0063] merge the nodes in the directed graph based on the similarity between the track groups, to obtain new nodes;

[0064] for each node currently contained in the directed graph, perform the following operations: if an out-degree of a node is greater than a first out-degree threshold, or an in-degree of a node is greater than a first in-degree threshold, determine the node as a first node, wherein the out-degree of the node is a number of directed edges with the node as a starting point, and the in-degree of the node is a number of directed edges with the node as a terminal point;

[0065] merge each first node in the obtained first node set that is connected to each other, into a corresponding second node.

[0066] Optionally, when the nodes in the directed graph are merged based on the similarity between the track groups to obtain the new nodes, the second processing module is further configured to:

[0067] for each track grid, perform the following operations: cluster each track group in a track grid and its eight neighboring track grids that have a similarity reaching a second similarity threshold and are not connected, to obtain at least one track group cluster;

[0068] merge each node in the directed graph corresponding to each track group belonging to a same track group cluster into a corresponding new node.

[0069] Optionally, based on the nodes in the directed graph and the directed edges between the nodes, the target map region corresponding road intersection skeleton, road skeleton and road network topology relationship are obtained, and the generation module is further configured to:

[0070] determine road nodes and intersection nodes in the directed graph based on the out-degree and the in-degree of each node in the directed graph;

[0071] generate the target map region corresponding intersection skeleton based on each track group corresponding to each intersection node;

[0072] generate the target map region corresponding road skeleton based on each track group corresponding to each road node;

[0073] generate a road network topology relationship corresponding to the target map area based on the connectivity relationship between the nodes.

[0074] Optionally, based on the out-degree and the in-degree of each node in the directed graph, the road nodes and the intersection nodes in the directed graph are determined, and the generation module is further configured to:

[0075] For each node in the directed graph, the following operations are performed respectively:

[0076] If the out-degree of a node is greater than a second out-degree threshold, or the in-degree of a node is greater than a second in-degree threshold, the node is determined to be an intersection node.

[0077] If the out-degree of a node is not greater than the second out-degree threshold, and the in-degree of a node is not greater than the second in-degree threshold, the node is determined to be a road node.

[0078] In a third aspect, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the program to implement the steps of the method of any one of the first aspect.

[0079] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the steps of the method of any one of the first aspect.

[0080] In the embodiments of the present application, the service device obtains each trajectory data corresponding to the target map area from the terminal device, divides each trajectory data into each trajectory group, and converts each trajectory group into a corresponding node in the directed graph. Based on the driving information of each trajectory data contained in each trajectory group, the connectivity relationship between each trajectory group is determined, and the connectivity relationship between two connected trajectory groups is converted into a directed edge between the corresponding two nodes in the directed graph. Based on each node in the directed graph and the directed edge between each node, the intersection skeleton, the road skeleton, and the road network topology relationship corresponding to the target map area are obtained. In this way, based on the massive trajectory data, the nodes and the directed edges of the directed graph are established, the intersection nodes and the road nodes are determined according to the number of directed edges of each node, and the intersection skeleton, the road skeleton, and the road network topology relationship corresponding to the target map area are generated. This can avoid errors caused by manual labeling and improve the accuracy of the road network skeleton generation. BRIEF DESCRIPTION OF DRAWINGS

[0081] Figure 1 This is an application scenario diagram in the embodiments of the present application;

[0082] Figure 2 This is a flowchart of a road network skeleton generation method in the embodiments of the present application;

[0083] Figure 3 Fig. 1 is a schematic diagram of converting trajectory groups into corresponding nodes in a directed graph in an embodiment of the present application;

[0084] Figure 4 Fig. 2 is a schematic diagram of a process of obtaining trajectory groups in an embodiment of the present application;

[0085] Figure 5 Fig. 3 is a schematic diagram of a sub-region grid set corresponding to a target map region in an embodiment of the present application;

[0086] Figure 6 Fig. 4 is a schematic diagram of a trajectory grid in an embodiment of the present application;

[0087] Figure 7 Fig. 5 is a schematic diagram of obtaining at least one target cluster corresponding to a trajectory grid in an embodiment of the present application;

[0088] Figure 8 Fig. 6 is a schematic diagram of obtaining trajectory groups in an embodiment of the present application;

[0089] Figure 9 Fig. 7 is a schematic diagram of a process of determining a connection relationship between trajectory groups in an embodiment of the present application;

[0090] Figure 10 Fig. 8 is a schematic diagram of determining a connection relationship between trajectory groups in an embodiment of the present application;

[0091] Figure 11 Fig. 9 is a schematic diagram of a process of determining a directed edge in a directed graph in an embodiment of the present application;

[0092] Figure 12 Fig. 10 is a schematic diagram of determining a directed edge in a directed graph in an embodiment of the present application;

[0093] Figure 13 Fig. 11 is a schematic diagram of a process of node optimization in an embodiment of the present application;

[0094] Figure 14 Fig. 12 is a schematic diagram of a process of obtaining new nodes in an embodiment of the present application;

[0095] Figure 15 Fig. 13 is a schematic diagram of obtaining at least one trajectory group cluster in an embodiment of the present application;

[0096] Figure 16 Fig. 14 is a schematic diagram of obtaining a second node in an embodiment of the present application;

[0097] Figure 17 Fig. 15 is a schematic diagram of a process of obtaining a road network skeleton in an embodiment of the present application;

[0098] Figure 18 Fig. 16 is a schematic diagram of a process of determining road nodes and intersection nodes in a directed graph in an embodiment of the present application;

[0099] Figure 19 Fig. 1 is a schematic diagram of determining road nodes and intersection nodes in a directed graph in an embodiment of the present application;

[0100] Figure 20 Fig. 2 is a schematic diagram of a road network skeleton in an embodiment of the present application;

[0101] Figure 21 Fig. 3 is a schematic diagram of a road network skeleton generation device in an embodiment of the present application;

[0102] Figure 22 Fig. 4 is a schematic diagram of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0103] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, and not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0104] Some terms in the embodiments of the present application are explained below to facilitate understanding by those of ordinary skill in the art.

[0105] (1) Directed graph: composed of a set of nodes and a set of directed edges, each directed edge connecting an ordered pair of nodes.

[0106] (2) Out-degree of a node: the number of directed edges with the node as the starting point.

[0107] (3) In-degree of a node: the number of directed edges with the node as the end point.

[0108] The preferred embodiments of the present application will be described in detail below with reference to the drawings.

[0109] As Figure 1As shown, it is an application scenario diagram of the embodiment of the present application. It is an application scenario diagram of the embodiment of the present application. In the application scenario diagram, it includes a service device 110 and terminal devices 120 (including a terminal device 1201, a terminal device 1202,..., and a terminal device 120n). The service device 110 receives each trajectory data sent by the terminal device 120, the service device 110 divides each trajectory data obtained into each trajectory group, and respectively converts each trajectory group into a corresponding node in a directed graph, then determines the connection relationship between each trajectory group based on the driving information of each trajectory data contained in each trajectory group, and converts the connection relationship between the two connected trajectory groups into a directed edge between the corresponding two nodes in the directed graph, and finally obtains the road network skeleton corresponding to the target map area based on each node in the directed graph and the directed edge between each node.

[0110] The service device 110 can be a stand-alone physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, content delivery network (CDN), and big data and artificial intelligence platform. The terminal device 120 and the service device 110 can be directly or indirectly connected through wired or wireless communication, which is not limited in the present application.

[0111] The terminal device 120 can be a mobile phone, a portable computer, etc. carried by a driver during driving a vehicle, or a device with positioning function arranged in the vehicle, which is not limited in the present application.

[0112] Based on the above embodiment, refer to Figure 2 As shown, it is a flow diagram of a road network skeleton generation method in the embodiment of the present application, which specifically includes:

[0113] Step 20: obtaining each trajectory data corresponding to the target map area.

[0114] Each trajectory data contains the driving trajectory of the corresponding vehicle in the target map area.

[0115] In the embodiment of the present application, the service device obtains each trajectory data corresponding to the target map area from the terminal device.

[0116] Step 21: dividing each trajectory data obtained into each trajectory group, and respectively converting each trajectory group into a corresponding node in a directed graph.

[0117] The similarity between each trajectory data contained in each trajectory group reaches a similarity threshold.

[0118] In this embodiment of the application, after obtaining the trajectory data, the trajectory data is divided into trajectory groups, and each trajectory group is converted into a corresponding node in the directed graph.

[0119] For example, see Figure 3 The diagram shown is a schematic of how a trajectory group is transformed into a corresponding node in a directed graph in an embodiment of this application. Assuming that the obtained trajectory groups include: trajectory group A, trajectory group B, and trajectory group C, then trajectory group A is transformed into node A in the directed graph, trajectory group B is transformed into node B in the directed graph, and trajectory group C is transformed into node C in the directed graph.

[0120] Specifically, in this embodiment of the application, when performing step 21, it is necessary to divide the obtained trajectory data into trajectory groups, as described in the following document. Figure 4 The diagram shown is a flowchart illustrating the process of obtaining each trajectory group in an embodiment of this application. The following is a detailed explanation in conjunction with the attached diagram. Figure 4 The specific operations to be performed will be explained in detail:

[0121] Step 210: Divide the target map area according to the preset area segmentation method to obtain the sub-area raster set corresponding to the target map area.

[0122] In this embodiment of the application, a reference point is selected as the origin in the target map area, a planar coordinate system is established, and the target map area is divided according to the preset grid size to obtain the sub-region grid set corresponding to the target map area.

[0123] The preset grid size can be determined according to the actual situation, and this application embodiment does not impose any restrictions on it.

[0124] For example, see Figure 5 As shown, this is a schematic diagram of the sub-region grid set corresponding to the target map area in an embodiment of this application. Assuming that the reference point selected in the target map area is the lower left corner and the grid size is 20m*20m, the lower left corner is taken as the origin to establish a planar coordinate system, and the target map area is divided into sub-region grids according to the 20m*20m grid.

[0125] Step 211: Filter out each trajectory grid from the sub-region grid set.

[0126] Each trajectory grid contains at least one segment of trajectory data.

[0127] In the embodiments of the present application, after obtaining the sub-region grid set, the following operations are performed for each sub-region grid: judging whether one sub-region grid contains a piece of trajectory data in at least one trajectory data, if the sub-region grid contains a piece of trajectory data in at least one trajectory data, the sub-region grid is taken as a trajectory grid.

[0128] For example, referring to FIG. 2, which is a schematic diagram of each trajectory grid in the embodiments of the present application, wherein the sub-region grid 1 does not contain a piece of trajectory data in one trajectory data, and the sub-region grid 2 contains a piece of trajectory data in multiple trajectory data, and the sub-region grid 2 is taken as a trajectory grid. Figure 6

[0129] Step 212: for each trajectory grid, the following operations are performed respectively: clustering each piece of trajectory data in a trajectory grid with a similarity reaching a first similarity threshold to obtain at least one target cluster.

[0130] In the embodiments of the present application, after screening each trajectory grid from the sub-region grid set, the following operations are performed for each trajectory grid: clustering based on the similarity between each piece of trajectory data in a trajectory grid, and aggregating each piece of trajectory data with a similarity reaching a first similarity threshold into a target cluster to obtain at least one target cluster corresponding to the trajectory grid.

[0131] For example, referring to FIG. 3, which is a schematic diagram of obtaining at least one target cluster corresponding to a trajectory grid in the embodiments of the present application, clustering based on the visual field similarity between each piece of trajectory data in a trajectory grid, the visual field similarity between trajectory data a and trajectory data c is 0.71, the visual field similarity between trajectory data a and trajectory data d is 0.73, and the visual field similarity between trajectory data c and trajectory data d is 0.72, and trajectory data a, c and d with a visual field similarity greater than a first similarity threshold 0.7 are aggregated into one target cluster. Figure 7

[0132] Among them, the trajectory orientation information, the observable visual field range and the camera coordinate information of each trajectory point of the trajectory data corresponding to the vehicle can be obtained through the camera device installed in the vehicle with shooting function, and the visual field similarity between two trajectory data is determined based on the trajectory orientation information, the observable visual field range and the camera coordinate information.

[0133] Step 213: each target cluster obtained is taken as a corresponding trajectory group.

[0134] In the embodiments of the present application, after obtaining each target cluster, the following operations are performed for each target cluster: taking one target cluster as a trajectory group.

[0135] For example, referring to FIG. 4, which is a schematic diagram of obtaining each trajectory group in the embodiments of the present application, taking each target cluster as a trajectory group.​​Figure 8 As shown in the figure, the target class cluster A is taken as a trajectory group A, the target class cluster B is taken as a trajectory group B, and the target class cluster C is taken as a trajectory group C.

[0136] Step 22: Determine the connection relationship between the trajectory groups based on the driving information of the trajectory data included in each trajectory group, and convert the connection relationship between the two connected trajectory groups into a directed edge between the corresponding two nodes in the directed graph.

[0137] The driving information of each trajectory data at least includes a driving time corresponding to the trajectory data.

[0138] Specifically, in the embodiment of the present application, when step 22 is performed, the connection relationship between the trajectory groups needs to be determined based on the driving information of the trajectory data included in each trajectory group. Referring to Figure 9 As shown in the figure, the target class cluster A is taken as a trajectory group A, the target class cluster B is taken as a trajectory group B, and the target class cluster C is taken as a trajectory group C. Figure 9 The specific operations performed are described in detail as follows:

[0139] Step 220: Determine whether the first trajectory data included in one trajectory group is continuous with the second trajectory data included in another trajectory group, if yes, execute step 221, otherwise, execute step 224.

[0140] Step 221: Determine whether the driving time corresponding to the second trajectory data is before the driving time corresponding to the first trajectory data, if yes, execute step 222, otherwise, execute step 223.

[0141] Step 222: Determine that the trajectory group corresponding to the second trajectory data is the predecessor connection of the trajectory group corresponding to the first trajectory data.

[0142] In the embodiment of the present application, whether the first trajectory data included in one trajectory group is continuous with the second trajectory data included in another trajectory group is determined, if the first trajectory data included in one trajectory group is continuous with the second trajectory data included in another trajectory group, whether the driving time corresponding to the second trajectory data is before the driving time corresponding to the first trajectory data is determined, if the driving time corresponding to the second trajectory data is before the driving time corresponding to the first trajectory data, it is determined that the trajectory group corresponding to the second trajectory data is the predecessor connection of the trajectory group corresponding to the first trajectory data.

[0143] The first trajectory data included in one trajectory group is any one of the trajectory data in the trajectory group, and the second trajectory data included in another trajectory group is any one of the trajectory data in the trajectory group.

[0144] For example, refer to Figure 10 shown, assume that the trajectory data a contained in the trajectory group A is continuous with the trajectory data f contained in the trajectory group B, the driving time of the trajectory data a is 11:40, the driving time of the trajectory data f is 11:39, the driving time corresponding to the trajectory data f is before the driving time corresponding to the trajectory data a, and it is determined that the trajectory group B corresponding to the trajectory data f is the predecessor connection of the trajectory group A corresponding to the trajectory data a.

[0145] The driving time corresponding to the trajectory data can be the average of the driving times of the trajectory points contained in the trajectory data, or the driving time of the starting trajectory point or the terminal trajectory point contained in the trajectory data, which is not limited in the embodiment of the application.

[0146] Step 223: determining that the trajectory group corresponding to the second trajectory data is the successor connection of the trajectory group corresponding to the first trajectory data.

[0147] In the embodiment of the application, it is judged whether the first trajectory data contained in one trajectory group is continuous with the second trajectory data contained in another trajectory group. If the first trajectory data contained in one trajectory group is continuous with the second trajectory data contained in another trajectory group, it is judged whether the driving time corresponding to the second trajectory data is before the driving time corresponding to the first trajectory data. If the driving time corresponding to the second trajectory data is after the driving time corresponding to the first trajectory data, it is determined that the trajectory group corresponding to the second trajectory data is the successor connection of the trajectory group corresponding to the first trajectory data.

[0148] For example, as shown in Figure 10 shown, assume that the trajectory data a contained in the trajectory group A is continuous with the trajectory data h contained in the trajectory group C, the driving time of the trajectory data a is 11:40, the driving time of the trajectory data h is 11:41, the driving time corresponding to the trajectory data h is after the driving time corresponding to the trajectory data a, and it is determined that the trajectory group C corresponding to the trajectory data h is the successor connection of the trajectory group A corresponding to the trajectory data a.

[0149] Step 224: determining that the two trajectory groups are not connected.

[0150] In the embodiment of the application, it is judged whether the first trajectory data contained in one trajectory group is continuous with the second trajectory data contained in another trajectory group. If the first trajectory data contained in one trajectory group is not continuous with the second trajectory data contained in another trajectory group, it is determined that the two trajectory groups are not connected.

[0151] For example, as shown in Figure 10As shown in FIG. 6, assuming that any one of the trajectory data contained in the trajectory group A is discontinuous with any one of the trajectory data contained in the trajectory group D, it is determined that the trajectory group A is not connected with the trajectory group D.

[0152] Specifically, in the embodiment of the present application, when step 22 is performed, the connection relationship between the two connected trajectory groups needs to be converted into the directed edge between the corresponding two nodes in the directed graph, referring to FIG. 7. Figure 11 As shown in FIG. 7, it is a flow diagram for determining the directed edge in the directed graph in the embodiment of the present application. The following will be combined with FIG. 8 to describe the embodiment of the present application. Figure 11 The specific operation performed will be described in detail:

[0153] For each trajectory group, the following operations are performed respectively:

[0154] Step 225: Taking one trajectory group as a first trajectory group, taking a trajectory group which is predecessor connected with the first trajectory group as a second trajectory group, and converting the connection relationship between the first trajectory group and the second trajectory group into a directed edge between the corresponding two nodes in the directed graph.

[0155] Wherein, the direction of the directed edge is from the node corresponding to the second trajectory group to the node corresponding to the first trajectory group.

[0156] In the embodiment of the present application, after obtaining the connection relationship between each trajectory group, one trajectory group is taken as a first trajectory group, a trajectory group which is predecessor connected with the first trajectory group is taken as a second trajectory group, and the connection relationship between the first trajectory group and the second trajectory group is converted into a directed edge between the node corresponding to the first trajectory group and the node corresponding to the second trajectory group in the directed graph.

[0157] For example, referring to FIG. 7, Figure 12 As shown in FIG. 7, it is a flow diagram for determining the directed edge in the directed graph in the embodiment of the present application. The following will be combined with FIG. 8 to describe the embodiment of the present application.

[0158] Step 226: Taking one trajectory group as a first trajectory group, taking a trajectory group which is successor connected with the first trajectory group as a third trajectory group, and converting the connection relationship between the first trajectory group and the third trajectory group into a directed edge between the corresponding two nodes in the directed graph.

[0159] Wherein, the direction of the directed edge is from the node corresponding to the first trajectory group to the node corresponding to the third trajectory group.

[0160] In the embodiments of the present application, after the connection relationship between each trajectory group is obtained, one trajectory group is taken as a first trajectory group, a trajectory group connected subsequently to the first trajectory group is taken as a third trajectory group, and the connection relationship between the first trajectory group and the third trajectory group is converted into a directed edge between a node corresponding to the first trajectory group and a node corresponding to the third trajectory group in the directed graph.

[0161] For example, as shown in FIG. 6, assuming that the trajectory group C is connected subsequently to the trajectory group A, the trajectory group A corresponds to a node A in the directed graph, the trajectory group C corresponds to a node C in the directed graph, and the connection relationship between the trajectory group A and the trajectory group C is converted into a directed edge between the node A and the node C in the directed graph, and the direction of the directed edge is from the node C to the node A. Figure 12

[0162] Further, in the embodiments of the present application, after the connection relationship between the two trajectory groups connected is converted into the directed edge between the two corresponding nodes in the directed graph, each node in the directed graph is optimized based on the similarity between each trajectory group, as shown in FIG. 7, which is a flowchart of the node optimization in the embodiments of the present application, and the following will be described in combination with FIG. 8, which is a flowchart of obtaining the new nodes in the embodiments of the present application. Figure 13 Figure 13 The specific operations performed in step 1300 will be described in detail as follows.

[0163] Step 1300: Each node in the directed graph is merged based on the similarity between each trajectory group, and new nodes are obtained.

[0164] Specifically, when step 1300 is performed, the service device specifically performs the following operations. As shown in FIG. 8, which is a flowchart of obtaining the new nodes in the embodiments of the present application, the following will be described in combination with FIG. 9, which is a flowchart of the node optimization in the embodiments of the present application. Figure 14 Figure 14 The specific operations performed in step 1300-1 will be described in detail as follows.

[0165] Step 1300-1: For each trajectory grid, the following operations are performed respectively: the trajectory groups in a trajectory grid and the trajectory grids in the eight neighborhoods of the trajectory grid which are similar to each other and not connected are clustered, and at least one trajectory group cluster is obtained.

[0166] In the embodiments of the present application, for each trajectory grid, the following operations are performed respectively: the trajectory groups in a trajectory grid and the trajectory grids in the eight neighborhoods of the trajectory grid which are similar to each other and not connected are clustered, and at least one trajectory group cluster is obtained.

[0167] For example, as shown in FIG. 10, assuming that the trajectory group C is connected subsequently to the trajectory group A, the trajectory group A corresponds to a node A in the directed graph, the trajectory group C corresponds to a node C in the directed graph, and the connection relationship between the trajectory group A and the trajectory group C is converted into a directed edge between the node A and the node C in the directed graph, and the direction of the directed edge is from the node C to the node A. Figure 15 ​​​As shown in the figure, eight-neighbor trajectory grids of trajectory grid 1 include: trajectory grid 2, trajectory grid 3, trajectory grid 4, trajectory grid 5, trajectory grid 6, trajectory grid 7, trajectory grid 8, and trajectory grid 9. Based on the visual field similarity between each trajectory group in trajectory grid 1 and each trajectory group in the eight-neighbor trajectory grids of trajectory grid 1, clustering is performed. The visual field similarity between trajectory group A in trajectory grid 1 and trajectory group D in trajectory grid 2 is 0.81, and they are not connected. The visual field similarity between trajectory group A in trajectory grid 1 and trajectory group E in trajectory grid 5 is 0.83, and they are not connected. The visual field similarity between trajectory group D in trajectory grid 2 and trajectory group E in trajectory grid 5 is 0.82, and they are not connected. Trajectory groups A, E, and F, which have a visual field similarity greater than the second similarity threshold value 0.8 and are not connected, are aggregated into one trajectory group cluster.

[0168] The visual field similarity between two trajectory groups can be determined by the visual field similarity between each trajectory data in the two trajectory groups, or by the average of the visual field similarity between each trajectory data included in the two trajectory groups. The present application does not limit this.

[0169] Step 1300-2: Merge each node corresponding to each trajectory group belonging to the same trajectory group cluster in the directed graph into a corresponding new node.

[0170] In the present application, after obtaining at least one trajectory group cluster, the following operations are performed for each trajectory group cluster: merging each node corresponding to each trajectory group in a trajectory group cluster to generate a new node in the directed graph.

[0171] For example, as shown in the figure, one trajectory group cluster includes trajectory group A, trajectory group B, and trajectory group C. Node A corresponding to trajectory group A, node E corresponding to trajectory group E, and node F corresponding to trajectory group F are merged into a new node. Figure 15

[0172] For each node currently included in the directed graph, the following operations are performed:

[0173] Step 1301: Determine whether the out-degree of a node is greater than a first out-degree threshold value, or whether the in-degree of a node is greater than a first in-degree threshold value. If so, perform step 1302.

[0174] The out-degree of a node is the number of directed edges with the node as the starting point, and the in-degree of a node is the number of directed edges with the node as the terminal point.

[0175] ​Step 1302: determining a node as a first node.

[0176] In the embodiment of the present application, it is determined whether the out-degree of a node is greater than a first out-degree threshold or whether the in-degree of a node is greater than a first in-degree threshold. If the out-degree of a node is greater than the first out-degree threshold or the in-degree of the node is greater than the first in-degree threshold, the node is determined as a first node.

[0177] In the embodiment of the present application, the first out-degree threshold can be 1 and the first in-degree threshold can be 1, which are not limited.

[0178] For example, assuming that the first out-degree threshold is 1 and the first in-degree threshold is 1, the out-degree of a node is 1, and the in-degree of a node is 2, the in-degree of the node is greater than the first in-degree threshold, and the node is determined as a first node.

[0179] Step 1303: merging each first node in the obtained first node set into a corresponding second node.

[0180] In the embodiment of the present application, each first node is taken as a first node set, and each first node in the first node set is merged to generate a corresponding second node.

[0181] For example, referring to FIG. 13B, which is a schematic diagram of obtaining a second node in the embodiment of the present application, it is assumed that the first node a and the first node b in the first node set are connected, the first node a and the first node c are connected, and the first node b and the first node c are connected. The first node a, the first node b, and the first node c are merged to generate a corresponding second node. Figure 16 In this way, based on the in-degree and the out-degree of each node, each node in the directed graph is merged, the optimization of the node is realized, and the accuracy of the road network skeleton can be improved.

[0182] Step 23: obtaining a road intersection skeleton, a road skeleton, and a road network topology relationship corresponding to a target map area based on each node in the directed graph and the directed edges between the nodes.

[0183] In the embodiment of the present application, each node in the directed graph and the directed edges between the nodes are converted into a road network skeleton corresponding to the target map area.

[0184] Specifically, when step 23 is performed, the service device specifically performs the following operations. Referring to FIG. 13C, which is a flowchart of obtaining a road network skeleton in the embodiment of the present application, the following will be described in combination with FIG. 13A and FIG. 13B.

[0185] Figure 17 The specific operations performed are described in detail as follows: Figure 17

[0186] ​​Step 230: determining the road node and the intersection node in the directed graph based on the out-degree and the in-degree of each node in the directed graph.

[0187] Specifically, when step 230 is performed, the service device specifically performs the following operations. Referring to Figure 18 As shown in the figure, it is a flowchart for determining the road node and the intersection node in the directed graph in the embodiment of the present application. The following will be described in combination with the Figure 18 The specifically performed operations will be described in detail as follows:

[0188] For each node in the directed graph, the following operations are performed respectively:

[0189] Step 2300: judging whether the out-degree of a node is greater than the second out-degree threshold or the in-degree of a node is greater than the second in-degree threshold, if yes, performing step 2301, otherwise, performing step 2302.

[0190] Step 2301: determining a node as an intersection node.

[0191] In the embodiment of the present application, if the out-degree of a node is greater than the second out-degree threshold or the in-degree of a node is greater than the second in-degree threshold, the node is determined as an intersection node.

[0192] The second out-degree threshold can be 1 and the second in-degree threshold can be 1, which are not limited in the embodiment of the present application.

[0193] For example, referring to Figure 19 As shown in the figure, it is a schematic diagram for determining the road node and the intersection node in the directed graph in the embodiment of the present application. It is assumed that the second out-degree threshold is 1, the second in-degree threshold is 1, the out-degree of node 1 is 4, and the in-degree of a node is 4. The in-degree of node 1 is greater than the second in-degree threshold, and the out-degree of node 1 is greater than the second out-degree threshold. Therefore, the node is determined as an intersection node.

[0194] Step 2302: determining a node as a road node.

[0195] In the embodiment of the present application, if the out-degree of a node is not greater than the second out-degree threshold and the in-degree of a node is not greater than the second in-degree threshold, the node is determined as a road node.

[0196] For example, as Figure 19As shown, assuming that the second out-degree threshold is 1, the second in-degree threshold is 1, the out-degree of node 2 is 1, and the in-degree of one node is 1, it is determined that the in-degree of node 2 is not greater than the second in-degree threshold, and the out-degree of node 2 is not greater than the second out-degree threshold, and node 2 is a road node.

[0197] Step 231: generating a road intersection skeleton corresponding to the target map region based on the trajectory groups corresponding to the road intersection nodes respectively.

[0198] In the embodiment of the present application, after obtaining the road intersection nodes, the trajectory groups corresponding to the road intersection nodes respectively are converted into the road intersection skeleton corresponding to the target map region.

[0199] For example, as shown in Figure 20 As shown, the figure is a schematic diagram of the road network skeleton in the embodiment of the present application, and the trajectory groups corresponding to road intersection node 1 are converted into the road intersection skeleton corresponding to the target map region.

[0200] Step 232: generating a road skeleton corresponding to the target map region based on the trajectory groups corresponding to the road nodes respectively.

[0201] In the embodiment of the present application, after obtaining the road nodes, the trajectory groups corresponding to the road nodes respectively are converted into the road skeleton corresponding to the target map region.

[0202] For example, as shown in Figure 20 As shown, the trajectory groups corresponding to road node 2 are converted into the road skeleton corresponding to the target map region.

[0203] Step 233: generating a road network topology relationship corresponding to the target map region based on the connectivity relationship corresponding to the directed edges between the nodes.

[0204] In the embodiment of the present application, the connectivity relationship corresponding to the directed edges between the nodes is converted into the road network topology relationship corresponding to the target map region.

[0205] Further, in the embodiment of the present application, after the road intersection skeleton, the road skeleton and the road network topology relationship corresponding to the target map region are generated, the division positions of the road skeleton and the road intersection skeleton can also be updated according to the change of the angle information of the trajectory data reaching different road intersections at the road intersection, so as to obtain more accurate road intersection boundaries.

[0206] Optionally, in the embodiment of the present application, more fine nodes can also be generated in the road skeleton and the road intersection skeleton according to equal intervals, so as to form new road skeleton and road intersection skeleton outputs.

[0207] Based on the same inventive concept, the embodiment of the present application also provides a road network skeleton generation device, as shown in Figure 21 As shown, the figure is a structural schematic diagram of the road network skeleton generation device in the embodiment of the present application, and the device specifically comprises:

[0208] The acquisition module 2101 is configured to acquire each trajectory data corresponding to a target map area, where each trajectory data includes a driving trajectory of a corresponding vehicle in the target map area.

[0209] The grouping module 2102 is configured to divide each trajectory data into each trajectory group, and convert each trajectory group into a corresponding node in a directed graph.

[0210] The first processing module 2103 is configured to determine a connection relationship between each trajectory group based on driving information of each trajectory data included in each trajectory group, and convert the connection relationship between two connected trajectory groups into a directed edge between two corresponding nodes in the directed graph.

[0211] The generation module 2104 is configured to obtain a road intersection skeleton, a road skeleton and a road network topology relationship corresponding to the target map area based on each node in the directed graph and the directed edge between each node.

[0212] Optionally, when each trajectory data is divided into each trajectory group, the grouping module 2102 is further configured to:

[0213] divide the target map area according to a preset area division manner to obtain a sub-region grid set corresponding to the target map area;

[0214] select each trajectory grid from the sub-region grid set, where each trajectory grid includes a section of trajectory data in at least one trajectory data;

[0215] for each trajectory grid, the following operations are performed: clustering each section of trajectory data in a trajectory grid, where the similarity of each section of trajectory data reaches a first similarity threshold, to obtain at least one target cluster;

[0216] respectively, each target cluster obtained is taken as a corresponding trajectory group.

[0217] Optionally, the driving information of each trajectory data at least includes a driving time corresponding to the corresponding trajectory data.

[0218] Then, when the connection relationship between each trajectory group is determined based on the driving information of each trajectory data included in each trajectory group, the first processing module 2103 is further configured to:

[0219] for each trajectory group, the following operations are performed:

[0220] If the first trajectory data contained in one trajectory group is continuous with the second trajectory data contained in another trajectory group, and the driving time corresponding to the second trajectory data is before the driving time corresponding to the first trajectory data, it is determined that the trajectory group corresponding to the second trajectory data is the predecessor of the trajectory group corresponding to the first trajectory data.

[0221] If the first trajectory data contained in one trajectory group is continuous with the second trajectory data contained in another trajectory group, and the driving time corresponding to the second trajectory data is after the driving time corresponding to the first trajectory data, it is determined that the trajectory group corresponding to the second trajectory data is the successor of the trajectory group corresponding to the first trajectory data.

[0222] Optionally, when the connection relationship between the two connected trajectory groups is converted into a directed edge between the corresponding two nodes in the directed graph, the first processing module 2103 is further used to:

[0223] For each trajectory group, the following operations are respectively performed:

[0224] One trajectory group is taken as a first trajectory group, a trajectory group which is the predecessor of the first trajectory group is taken as a second trajectory group, and the connection relationship between the first trajectory group and the second trajectory group is converted into a directed edge between the corresponding two nodes in the directed graph, wherein the direction of the directed edge is from the node corresponding to the second trajectory group to the node corresponding to the first trajectory group.

[0225] One trajectory group is taken as a first trajectory group, a trajectory group which is the successor of the first trajectory group is taken as a third trajectory group, and the connection relationship between the first trajectory group and the third trajectory group is converted into a directed edge between the corresponding two nodes in the directed graph, wherein the direction of the directed edge is from the node corresponding to the first trajectory group to the node corresponding to the third trajectory group.

[0226] Optionally, after the connection relationship between the two connected trajectory groups is converted into a directed edge between the corresponding two nodes in the directed graph, the apparatus further comprises a second processing module, and the second processing module 2105 is used to:

[0227] Based on the similarity between each trajectory group, each node in the directed graph is merged to obtain new each node;

[0228] For each node currently contained in the directed graph, the following operations are respectively performed: if the out-degree of one node is greater than a first out-degree threshold, or the in-degree of one node is greater than a first in-degree threshold, it is determined that one node is a first node, wherein the out-degree of one node is the number of directed edges with one node as the starting point, and the in-degree of one node is the number of directed edges with one node as the ending point.

[0229] Each first node in the obtained first node set is merged into a corresponding second node.

[0230] Optionally, based on the similarity between each trajectory group, each node in the directed graph is merged to obtain each new node, and the second processing module 2105 is further configured to:

[0231] For each trajectory grid, the following operations are respectively performed: clustering each trajectory group in one trajectory grid and its eight neighborhood trajectory grids which have a similarity reaching a second similarity threshold and are not connected to obtain at least one trajectory group cluster.

[0232] Each node corresponding to each trajectory group belonging to the same trajectory group cluster in the directed graph is merged into a corresponding new node.

[0233] Optionally, based on each node in the directed graph and the directed edges between each node, the road skeleton, the road skeleton and the road network topological relationship corresponding to the target map area are obtained, and the generation module 2104 is further configured to:

[0234] Based on the out-degree and the in-degree of each node in the directed graph, the road node and the intersection node in the directed graph are determined.

[0235] Based on each trajectory group corresponding to each intersection node, the intersection skeleton corresponding to the target map area is generated.

[0236] Based on each trajectory group corresponding to each road node, the road skeleton corresponding to the target map area is generated.

[0237] Based on the connected relationship corresponding to the directed edges between each node, the road network topological relationship corresponding to the target map area is generated.

[0238] Optionally, based on the out-degree and the in-degree of each node in the directed graph, the road node and the intersection node in the directed graph are determined, and the generation module 2104 is further configured to:

[0239] For each node in the directed graph, the following operations are respectively performed:

[0240] If the out-degree of a node is greater than a second out-degree threshold, or the in-degree of a node is greater than a second in-degree threshold, the node is determined to be an intersection node.

[0241] If the out-degree of a node is not greater than the second out-degree threshold, and the in-degree of a node is not greater than the second in-degree threshold, the node is determined to be a road node.

[0242] Based on the above embodiments, refer to Figure 22 FIG. 1 shows a structural schematic diagram of an electronic device according to an embodiment of the present application.

[0243] The electronic device can include a processor 2210 (Center Processing Unit, CPU), a memory 2220, an input device 2230, and an output device 2240, etc. The input device 2230 can include a keyboard, a mouse, a touch screen, etc. The output device 2240 can include a display device, such as a Liquid Crystal Display (LCD), a Cathode Ray Tube (CRT), etc.

[0244] The memory 2220 can include a Read-Only Memory (ROM) and a Random Access Memory (RAM), and provide the processor 2210 with program instructions and data stored in the memory 2220. In the embodiments of the present application, the memory 2220 can be used to store the programs of any of the road network skeleton generation methods in the embodiments of the present application.

[0245] The processor 2210 calls the program instructions stored in the memory 2220, and the processor 2210 is configured to execute any of the road network skeleton generation methods in the embodiments of the present application according to the obtained program instructions.

[0246] Based on the above embodiments, in the embodiments of the present application, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program. When the computer program is executed by a processor, the road network skeleton generation method in any of the method embodiments is implemented.

[0247] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can be in 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.

[0248] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general purpose computer, a special purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The functions specified in one or more flows and / or blocks

[0249] 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 Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0250] 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 that are executed on the computer or other programmable apparatus provide steps for implementing the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0251] 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 road network skeleton generation method characterized by comprising: The method comprises the following steps: obtaining each trajectory data corresponding to a target map area, wherein each trajectory data comprises a driving trajectory of a corresponding vehicle in the target map area; dividing each trajectory data into each trajectory group, and converting each trajectory group into a corresponding node in a directed graph, wherein the similarity between each trajectory data contained in each trajectory group reaches a similarity threshold; determining the connection relationship between each trajectory group based on the driving information of each trajectory data contained in each trajectory group, and converting the connection relationship between two connected trajectory groups into a directed edge between the corresponding two nodes in the directed graph; merging each node in the directed graph based on the similarity between each trajectory group, and obtaining new each node; for each node currently contained in the directed graph, the following operations are performed: if the out-degree of a node is greater than a first out-degree threshold, or the in-degree of the node is greater than a first in-degree threshold, the node is determined as a first node, wherein the out-degree of the node is the number of directed edges with the starting point of the node, and the in-degree of the node is the number of directed edges with the node as the terminal point; merging each first node in the obtained first node set into a corresponding second node; obtaining the intersection skeleton, the road skeleton and the road network topology relationship of the target map area based on each node in the directed graph and the directed edge between each node.

2. The method of claim 1, wherein, The method comprises the following steps: dividing the target map area according to a preset region segmentation method to obtain a sub-region grid set corresponding to the target map area; selecting each trajectory grid from the sub-region grid set, wherein each trajectory grid contains at least one segment of trajectory data in each trajectory data; for each trajectory grid, the following operations are performed: clustering each segment of trajectory data in a trajectory grid whose similarity reaches a first similarity threshold to obtain at least one target cluster; each obtained target cluster is taken as a corresponding trajectory group.

3. The method of claim 1 or 2, wherein, The driving information of each trajectory data at least comprises a driving time corresponding to the trajectory data; The method comprises the following steps: for each trajectory group, the following operations are performed: if the first trajectory data contained in one trajectory group is continuous with the second trajectory data contained in another trajectory group, and the driving time of the second trajectory data is before the driving time of the first trajectory data, it is determined that the trajectory group corresponding to the second trajectory data is the predecessor connection of the trajectory group corresponding to the first trajectory data; if the first trajectory data contained in one trajectory group is continuous with the second trajectory data contained in another trajectory group, and the driving time of the second trajectory data is after the driving time of the first trajectory data, it is determined that the trajectory group corresponding to the second trajectory data is the successor connection of the trajectory group corresponding to the first trajectory data.

4. The method of claim 3, wherein, The converting of the connectivity between the two connected track groups into a directed edge between the corresponding two nodes in the directed graph comprises: For each track group, the following operations are performed respectively: Taking one track group as a first track group, taking a track group connected in front of the first track group as a second track group, and converting the connectivity between the first track group and the second track group into a directed edge between the corresponding two nodes in the directed graph, wherein the direction of the directed edge is from the node corresponding to the second track group to the node corresponding to the first track group; Taking one track group as a first track group, taking a track group connected in front of the first track group as a third track group, and converting the connectivity between the first track group and the third track group into a directed edge between the corresponding two nodes in the directed graph, wherein the direction of the directed edge is from the node corresponding to the first track group to the node corresponding to the third track group.

5. The method of claim 1, wherein, The merging of each node in the directed graph based on the similarity between the track groups comprises: For each track grid, the following operations are performed respectively: clustering each track group in one track grid and its eight neighborhood track grids which have a similarity reaching a second similarity threshold and are not connected, to obtain at least one track group cluster; Merging each node in the directed graph corresponding to each track group belonging to the same track group cluster into a corresponding new node.

6. The method of claim 1 or 2, wherein, The obtaining of the intersection skeleton, the road skeleton and the road network topology relationship corresponding to the target map area based on each node in the directed graph and the directed edges between the nodes comprises: Determining the road nodes and the intersection nodes in the directed graph based on the out-degree and the in-degree of each node in the directed graph; Generating the intersection skeleton corresponding to the target map area based on each track group corresponding to each intersection node; Generating the road skeleton corresponding to the target map area based on each track group corresponding to each road node; Generating the road network topology relationship corresponding to the target map area based on the connectivity corresponding to the directed edges between the nodes.

7. The method of claim 6, wherein, The determining of the road nodes and the intersection nodes in the directed graph based on the out-degree and the in-degree of each node in the directed graph comprises: For each node in the directed graph, the following operations are performed respectively: If the out-degree of a node is greater than a second out-degree threshold, or the in-degree of the node is greater than a second in-degree threshold, the node is determined as an intersection node; If the out-degree of a node is not greater than the second out-degree threshold, and the in-degree of the node is not greater than the second in-degree threshold, the node is determined as a road node.

8. A road network skeleton generation device characterized by comprising: The method comprises: An acquisition module is configured to acquire each track data corresponding to a target map area, wherein each track data contains a driving track of a corresponding vehicle in the target map area; A grouping module is configured to divide each track data obtained into each track group, and convert each track group into a corresponding node in a directed graph; wherein the similarity between each track data contained in each track group reaches a similarity threshold. The method comprises: An acquisition module is configured to acquire each track data corresponding to a target map area, wherein each track data contains a driving track of a corresponding vehicle in the target map area; A grouping module is configured to divide each track data obtained into each track group, and convert each track group into a corresponding node in a directed graph; wherein the similarity between each track data contained in each track group reaches a similarity threshold. The first processing module is configured to determine a connection relationship between the trajectory groups based on driving information of the trajectory data included in the trajectory groups, and convert the connection relationship between two connected trajectory groups into a directed edge between two corresponding nodes in the directed graph. The second processing module is configured to merge the nodes in the directed graph based on the similarity between the trajectory groups, to obtain new nodes. For each node currently included in the directed graph, the following operations are performed: if the out-degree of a node is greater than a first out-degree threshold, or the in-degree of the node is greater than a first in-degree threshold, the node is determined as a first node, wherein the out-degree of the node is the number of directed edges with a starting point of the node, and the in-degree of the node is the number of directed edges with a terminal point of the node. Each first node in the obtained first node set that is connected is merged into a corresponding second node. The generation module is configured to obtain a road intersection skeleton, a road skeleton, and a road network topology relationship corresponding to the target map region based on the nodes in the directed graph and the directed edges between the nodes.

9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor implements the steps of the method in any one of claims 1-7 when executing the program.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program implements the steps of the method in any one of claims 1-7 when executed by the processor.

Images