Road network generation method, road network update system, and program product

Abstract

The invention provides a road network generation method, a road network updating system and a computer program product, and relates to the technical field of computers, the method comprises the following steps: obtaining layered geometric data of an indoor space, and constructing a graph structure of a road area in the indoor space according to the layered geometric data to determine a basic road network; nodes of the vertical traffic facilities are determined, and connecting roads of the nodes and a basic road network are generated, so that a connecting road network of the vertical traffic facilities and the basic road network is constructed; and carrying out main road extraction on a combined road network formed by the basic road network and the connection road network to determine an initial layer inner road network, optimizing the initial layer inner road network to obtain an layer inner road network, and carrying out interlayer communication on the layer inner road network to obtain an integral indoor road network. According to the invention, the whole indoor road network can be accurately generated.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of computer, in particular, to a road network generation method, a road network updating system and a computer program product. BACKGROUND

[0002] In scenarios such as analyzing regional road conditions and generating navigation maps, road network extraction is usually needed, and then subsequent analysis and processing are performed based on the extracted road network.

[0003] The road network extraction can be based on data collected by a laser radar and other sensors, or can also be based on an indoor building distribution map.

[0004] However, the method based on the laser radar and other sensors has a high cost of road network extraction and cannot be widely applied. The method based on the indoor building distribution map needs manual road network drawing according to the indoor building distribution map, and therefore a large amount of manpower and time needs to be invested, the operation efficiency is low, and the accuracy is poor. SUMMARY

[0005] The present disclosure aims to provide a road network generation method, a road network updating system and a computer program product, thereby at least partially overcoming the problem of low accuracy and low operation efficiency of road network generation caused by the limitations and defects of the related art.

[0006] Other characteristics and advantages of the present disclosure will become apparent from the following detailed description, or will be learned by practice of the present disclosure.

[0007] According to one aspect of the present disclosure, a road network generation method is provided, comprising: obtaining layered geometry data of an indoor space, constructing a graph structure of a road region in the indoor space according to the layered geometry data to determine a basic road network; determining nodes for vertical transportation facilities and generating connection roads of the nodes and the basic road network to construct a connection road network of the vertical transportation facilities and the basic road network; performing main road extraction on a combined road network composed of the basic road network and the connection road network to determine an initial intra-layer road network, optimizing the initial intra-layer road network to obtain an intra-layer road network, and performing inter-layer connectivity on the intra-layer road network to obtain an overall indoor road network.

[0008] In an exemplary embodiment of the present disclosure, the constructing a graph structure of a road region in the indoor space according to the layered geometry data to determine a basic road network comprises: performing binary image feature extraction on the layered geometry data to determine image features; Based on the image features, morphological skeleton extraction is performed, refining the road region in the image features into a center line with a single pixel width, while maintaining the topological connectivity of the road region, in order to determine the skeleton line; Key nodes are identified on the skeleton line, and the pixel sequence contained in each edge of the path segment connecting each key node is extracted. A graph structure containing the node coordinates and road coordinates of the key nodes is constructed to determine the basic road network.

[0009] In one exemplary embodiment of this disclosure, determining nodes for each vertical transportation facility and generating a connection network between the nodes and the underlying road network includes: Using the center point of the vertical transportation facility as a node, and taking the node as the starting point, search for the road closest to the starting point; Based on the road, determine the connecting road between the node and the basic road network, and obtain the connecting road network based on the connecting road.

[0010] In an exemplary embodiment of this disclosure, the step of extracting arterial roads from the combined road network formed by the basic road network and the connecting road network to determine the initial intra-layer road network includes: Convert the combined road network into an undirected graph; The undirected graph contains a loop structure, and adding a point will interrupt the loop structure; If there are edges of the first type in the undirected graph after the iterative query is interrupted, delete the edges of the first type until there are no edges of the first type in the undirected graph, so as to determine the first undirected graph; For each edge in the first undirected graph, determine the width data, iteratively determine the existence of edges of the second type based on the width data, delete the edges of the second type, until there are no edges of the second type in the undirected graph, so as to obtain the second undirected graph; The edges connected to nodes of degree 2 in the second undirected graph are merged to generate the initial intra-layer road network.

[0011] In one exemplary embodiment of this disclosure, optimizing the initial intra-layer road network to obtain the intra-layer road network includes: The original paths in the road network within the initial layer are geometrically simplified, and the simplified original paths are sampled to determine the sampling nodes; Establish topological relationships between nodes for each sampling node in the initial layer road network, and construct a topological graph based on the topological relationships of the initial layer road network; A distance field is constructed to determine spatial constraints and multiple constraint conditions. Based on the spatial constraints and multiple constraint conditions, the topology graph is optimized by nonlinear least squares solution to determine the intra-layer road network.

[0012] In one exemplary embodiment of this disclosure, determining the overall indoor road network by performing inter-layer connectivity on the intra-layer road network includes: The passage relationships between multiple vertical transportation facilities at different levels are parsed from the reference passage relationships stored in the first database; Based on the traffic relationship, the type of the vertical transportation facility is determined to be an elevator, and a two-way traffic edge is constructed between the nodes of multiple vertical transportation facilities; Based on the traffic relationship, the type of the vertical transportation facility is determined to be an escalator, and a one-way traffic edge is constructed between the nodes of multiple vertical transportation facilities; The traffic relationships of the vertical transportation facilities are vectorized into roads, and the nodes of the vertical transportation facilities are connected by bidirectional or unidirectional traffic edges to obtain the overall indoor road network.

[0013] According to one aspect of this disclosure, a road network update system is provided, comprising: The first database is used to store the current and historical distribution data of indoor spaces; The server, which is connected to the first database, is used to generate an overall indoor road network when it is determined that an overall indoor road network needs to be generated based on the difference results between the current distribution data and the historical distribution data, and to update the online map road network data of the indoor space based on the overall indoor road network; the overall indoor road network is obtained according to the above road network generation method. The second database is used to store updated online map road network data; The client is used to retrieve updated online map road network data from the second database for display.

[0014] In one exemplary embodiment of this disclosure, the server further includes: The connection relationship determination module is used to acquire the outdoor road network and determine the connection relationship between the overall indoor road network and the outdoor road network based on the door location information of the indoor space. The update control module is used to update the original overall indoor space of the indoor space according to the overall indoor road network, and after determining the connection relationship, update the online map road network data of the indoor space in the second database according to the overall indoor road network.

[0015] In one exemplary embodiment of this disclosure, the update control module includes: The difference result determination module is used to obtain the difference result between the current distribution data and the historical distribution data of the indoor space; The generation module is used to determine the overall indoor road network that needs to be generated based on the difference results, and to update the original overall indoor road network of the indoor space based on the overall indoor road network. The deletion module is used to determine the overall indoor road network that needs to be deleted based on the difference results, determine the update list of the indoor space, obtain the identification information of the update objects in the update list from the database, and perform a deletion operation on the original overall indoor road network of the indoor space through the identification information to update the overall indoor road network.

[0016] According to one aspect of this disclosure, a computer program product is provided, which, when executed by a processor, implements the road network generation method of any of the above.

[0017] The technical solution provided in this disclosure, on the one hand, involves acquiring layered geometric data of the indoor space and constructing a graph structure of road areas within the indoor space based on the layered geometric data to determine the basic road network; determining nodes for vertical transportation facilities and constructing a connecting road network between the vertical transportation facilities and the basic road network; extracting main roads based on the combined road network formed by the connecting road network and the basic road network to determine the initial intra-layer road network; optimizing the initial intra-layer road network and performing inter-layer connectivity within the intra-layer road network to obtain the overall indoor road network. This allows for the automatic acquisition of the connecting road network between vertical transportation facilities and the basic road network based on the layered geometric data, thereby automatically determining the overall indoor road network, improving the operational efficiency of road network generation and the accuracy of the overall indoor road network. On the other hand, since the connecting road network of vertical transportation facilities can be automatically acquired, the accessibility of important facilities is involved, achieving comprehensive road network coverage, reducing costs, and increasing versatility.

[0018] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0020] Figure 1 A schematic flowchart illustrating the road network generation method applicable to embodiments of this disclosure is shown.

[0021] Figure 2 A schematic diagram of layered geometric data is shown in an embodiment of this disclosure.

[0022] Figure 3 This illustration shows a flowchart of the process for determining the basic road network in an embodiment of this disclosure.

[0023] Figure 4The flowchart illustrating the generation of the overall indoor road network in an embodiment of this disclosure is shown schematically.

[0024] Figure 5 A schematic diagram illustrating image features in an embodiment of this disclosure is shown.

[0025] Figure 6 The diagram illustrates the generation of an intra-layer road network in an embodiment of this disclosure.

[0026] Figure 7 This schematic diagram illustrates how an embodiment of the present disclosure connects the intra-floor road network into a unified indoor road network.

[0027] Figure 8 This schematically illustrates an automated generation and updating system for the overall indoor road network corresponding to a shopping mall in an embodiment of this disclosure.

[0028] Figure 9 The diagram illustrates a block diagram of a road network update system according to an embodiment of the present disclosure. Detailed Implementation

[0029] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make this disclosure more comprehensive and complete, and to fully convey the concept of the example embodiments to those skilled in the art. The described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced with one or more of the specific details omitted, or other methods, components, apparatus, steps, etc., can be employed. In other instances, well-known technical solutions are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0030] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0031] In related technologies, road network extraction can be based on data collected by LiDAR and other sensors, or it can be based on indoor building distribution maps. However, the method based on LiDAR and other sensors is costly and cannot be widely used. The method based on indoor building distribution maps requires manual road network drawing based on the indoor building distribution maps, which requires a lot of manpower and time, resulting in low operational efficiency and poor accuracy.

[0032] To address the aforementioned technical issues, this disclosure provides a road network generation method that can be used to generate an overall indoor road network for any type of indoor space, thereby updating the overall indoor road network and enabling navigation or other path planning functions.

[0033] Next, refer to Figure 1 As shown, each step in the road network generation method in the embodiments of this disclosure will be described in detail.

[0034] In step S110, the layered geometric data of the indoor space is obtained, and the graph structure of the road area in the indoor space is constructed based on the layered geometric data to determine the basic road network.

[0035] In this embodiment, the indoor space can be the indoor space of any building, such as a shopping mall, hospital, supermarket, school building, library, etc. Here, a shopping mall is used as an example. When it is detected that an overall indoor road network needs to be generated, layered geometric data of the indoor space can be obtained to generate it. The layered geometric data can be obtained from distribution data. Distribution data can be the location data and access data of floors, shops, doors, elevators, and escalators within the indoor space. Location data includes geometry, floor arrangement, and the relationship between shops and floors; access data can include the access relationships of elevators and escalators. Specifically, the difference between the current distribution data and historical distribution data of the indoor space can be obtained, and the need to generate an overall indoor road network can be determined based on the difference results.

[0036] Current distribution data refers to the distribution data of the indoor space within the current period, which can be defined as a daily period. The current period can be the current day, specifically day T. Historical distribution data can be the distribution data from the previous period, which can be the day before. For example, current distribution data can be represented as the distribution data for day T, while historical distribution data can be the distribution data for day T-1. Current distribution data can be the latest data in the first database, while historical distribution data can be online data. The first database can be a database used to store the data.

[0037] The distribution data in the first database is collected from the indoor space's navigation map, specifically partitioned by date, with the latest daily distribution data fully stored in the first database. Since the current and historical distribution data are determined based on the indoor space's navigation map, the need for real-time image acquisition is avoided, reducing data acquisition costs, improving data acquisition efficiency, and also enhancing data quality.

[0038] After obtaining the current and historical distribution data of the indoor space, data preprocessing can be performed. This preprocessing includes data clustering, data binding, and coordinate system transformation. Specifically, the current and historical distribution data are clustered according to the entity objects in the indoor space; the floors are arranged from low to high, and shops, doors, elevators, escalators, and their corresponding floors are bound together to construct a table of access relationships between elevators, escalators, and stairs; finally, the current and historical distribution data are transformed from a latitude and longitude coordinate system to a meter coordinate system.

[0039] Furthermore, the difference between the preprocessed current distribution data and the preprocessed historical distribution data can be calculated. For example, a difference operation can be performed on the current distribution data represented by day T and the historical distribution data represented by day T-1 to obtain a difference result. The difference result can be used to evaluate whether to generate or delete the overall indoor road network. The generation of the overall indoor road network can be triggered when the current distribution data changes or is added compared to the historical distribution data. The deletion of the overall indoor road network can be triggered when the current distribution data is deleted or changes compared to the historical distribution data. For example, when the difference result shows that the current distribution data on day T has new data compared to the historical distribution data on day T-1, it indicates that a new shopping mall has been added to the database, and the overall indoor road network needs to be generated; when the difference result shows that the current distribution data on day T has deleted data compared to the historical distribution data on day T-1, it indicates that a shopping mall has closed, and the overall indoor road network needs to be deleted accordingly; when the current distribution data on day T has data changes at the single shopping mall level compared to the historical distribution data on day T-1, it indicates that the shopping mall has changes such as the opening or closing of floors, shops, elevators and escalators, etc., and the overall indoor road network needs to be deleted and generated, thereby updating the overall indoor road network.

[0040] Based on the difference between the current distribution data and historical distribution data of the indoor space, if it is determined that an overall indoor road network needs to be generated, an overall indoor road network generation algorithm can be used to generate the network. For example, the distribution data of the indoor space is first broken down to extract its layered structure, resulting in layered geometric data. Specifically, a shopping mall can be divided into n layers, each layer containing one floor-level geometry, multiple shop front geometries, multiple escalator front geometries, and multiple doorway geometries, to obtain layered geometric data. This layered geometric data can be represented as an image. Figure 2 As shown, the layered geometric data may include floor surface geometry 201, shop surface geometry 202, escalator surface geometry 203, elevator surface geometry 204, and door point 205.

[0041] After obtaining the layered geometric data, feature extraction can be performed on the layered geometric data to obtain image features. For example, binary image feature extraction is performed from the layered geometric data of the floors to obtain image features. Among them, the floor surface is represented by a pixel value of 255 as white; the shop surface, elevator surface, and escalator surface are represented by a pixel value of 128 as gray; and irrelevant areas are represented by a pixel value of 0 as black, where one pixel represents an actual physical distance of 0.2m.

[0042] After obtaining the image features, road skeleton extraction can be performed. For example, a graph structure of road regions in an indoor space can be constructed based on the image features to determine the basic road network. Figure 3 The flowchart illustrating the determination of the basic road network is shown in the figure. (Refer to...) Figure 3 As shown, the main steps include: Step S310: Perform binary image feature extraction on the layered geometric data to determine image features; Step S320: Based on image features, perform morphological skeleton extraction, refine the road region in the image features into a center line with a width of one pixel, and maintain the topological connectivity of the road region to determine the skeleton line; Step S330: Identify key nodes on the skeleton line, extract the pixel sequence contained in each edge of the path segment connecting each key node, construct a graph structure containing the node coordinates and road coordinates of the key nodes, and determine the basic road network.

[0043] In this embodiment of the disclosure, the obtained image features are used as input, and a morphological skeleton extraction algorithm is used to refine the road region into a center line with a width of one pixel, while maintaining the topological connection relationship of the road, so as to obtain the skeleton line of the road region.

[0044] After obtaining the skeleton lines of the road region, a graph structure can be constructed, which can then be used to define the basic road network. For example, key nodes can be identified on the skeleton lines; these key nodes may include endpoints and intersections. Further, the pixel sequence of each edge in the path segments connecting each key node is extracted. Based on this pixel sequence, a graph structure containing the node coordinates and road coordinates of the key nodes is constructed, and this graph structure can be used to define the basic road network. The basic road network refers to the road network of the road region.

[0045] In this embodiment, the system automatically identifies three states—data addition, data deletion, and data change—based on the difference results between the current distribution data and historical distribution data, and triggers corresponding road network update strategies according to the difference results. Since only the changed portions are processed, a full regeneration is avoided, reducing computational load and costs, and improving road network update efficiency.

[0046] In step S120, nodes of the vertical transportation facilities are determined and connecting roads between the nodes and the basic road network are generated to construct the connecting road network between the vertical transportation facilities and the basic road network.

[0047] In this embodiment of the disclosure, vertical transportation facilities refer to automated devices that allow passage outside of road areas, specifically including elevators and escalators capable of automatic vertical movement between floors. The number of vertical transportation facilities can be one or more, determined according to actual needs. Vertical transportation facilities can be elevators and escalators within indoor spaces.

[0048] For example, for each vertical transportation facility, the center point of the facility can be obtained, and this center point can be used as a node. Starting from the node, the nearest road can be searched. After the road is determined, connecting roads between the node and the basic road network of the road area can be determined based on the road. All connecting roads are then combined to obtain a connecting road network. During the generation of connecting roads, it is necessary to ensure that the connecting roads do not cross other surface structures.

[0049] Specifically, the center point is marked as the connection starting node, and an R-tree index is constructed for the basic road network. Each road segment in the basic road network is treated as a rectangular boundary and organized hierarchically into the R-tree. An initial search radius is set with the connection starting node as the center. The R-tree index is used to quickly filter out road segments whose boundaries intersect with the search range corresponding to this initial search radius, forming a candidate road segment set. If there are no candidate road segments within the initial radius, the radius is automatically expanded until at least one candidate road segment is selected.

[0050] For the candidate road segment set, calculate the perpendicular distance from the starting node to each road segment. Sort all candidate road segments in ascending order of distance, select the road segment with the smallest perpendicular distance, and record the point on that road segment closest to the starting node as a candidate endpoint point. Generate an initial connecting road based on the distance from the starting node to the candidate endpoint point. If the initial connecting road does not conflict with other surface structures, use the initial connecting road as the connecting road. If the initial connecting road conflicts with other surface structures, update the initial connecting road and use the updated connecting road as the connecting road.

[0051] By adding fixed POI reachable paths as connecting roads, the road network coverage is increased, enabling path planning from any location to elevators and escalators, thus improving navigation integrity.

[0052] In step S130, the main roads of the combined road network consisting of the basic road network and the connecting road network are extracted to determine the initial intra-layer road network. The initial intra-layer road network is optimized to obtain the intra-layer road network. The intra-layer road network is then connected between layers to obtain the overall indoor road network.

[0053] In this embodiment, after obtaining the combined road network consisting of the basic road network and the connecting road network, the combined road network can be represented by an undirected graph. Iterative pruning, road filtering, and edge merging are then performed on the undirected graph representing the combined road network, and inter-layer relationships are imported to obtain the overall indoor road network. For example, firstly, it can be determined whether a loop structure exists in the undirected graph. If a loop structure exists, a point can be added to the combined road network to interrupt the loop structure. Further, iterative queries are performed to check whether a first type of edge exists in the undirected graph after the interruption. The first type of edge refers to an edge connecting 1-degree nodes of non-vertical transportation facilities with a length less than a first value, which can be 10m or other suitable values. For example, a first type of edge can be an edge connecting 1-degree nodes of non-elevator and escalator facilities with a length less than 10m. If a first type of edge exists in the undirected graph after the interruption, iterative queries can be stopped until no first type of edge exists in the undirected graph after the interruption, thus completing iterative pruning. The undirected graph with the first type of edge deleted can be used as the first undirected graph.

[0054] Iterative pruning can eliminate disordered branches generated by the skeleton algorithm, remove unnecessary detailed branches, retain key connection points such as escalators and elevators, and improve the quality and integrity of the road network.

[0055] Due to the low precision of the basic geometric data, narrow gaps may exist between shop fronts, which may also be extracted as roads by the morphological skeleton extraction algorithm. To avoid these gaps being incorrectly extracted as roads, road filtering can be performed. For example, the width data of each edge in the first undirected graph is determined. Based on the width data, iterative determination is made to identify edges of the second type in the first undirected graph. Edges of the second type are then deleted until no edges of the second type remain in the first undirected graph, thus obtaining the second undirected graph. When determining the width data, it can be set according to the road width in the binary graph. Here, the second type of edge refers to an edge whose width data is less than a second value and whose deletion does not affect the global connectivity of the first undirected graph. The second value can be, for example, 2 meters, or other values ​​depending on actual needs. Based on this, the second type of edge can be an edge whose width data is less than 2 meters and whose deletion does not affect the global connectivity of the first undirected graph. Edges of the second type can be deleted, and the iteration of deleting second type edges stops when no edges of the second type remain in the first undirected graph, thus obtaining the second undirected graph.

[0056] By filtering out impassable narrow roads, network integrity is ensured, global connectivity of the road network is maintained, and navigation usability and accuracy are improved.

[0057] Furthermore, edge merging can be performed. For example, edges connecting second-degree nodes in the second undirected graph can be merged to generate the overall indoor road network. A second-degree node is a node connecting two edges. To simplify the second undirected graph, edges connecting nodes that are second-degree nodes can be merged into a single edge, thus achieving edge merging and obtaining the initial intra-layer road network. The initial intra-layer road network refers to the road network of each floor in the indoor space, and the initial intra-layer road network of each floor can be different. It should be noted that if merging second-degree nodes results in a cycle, the merging of the second-degree nodes should be cancelled.

[0058] After obtaining the initial intra-layer road network, which is already a simplified representation of the indoor road network, the results obtained by the morphological skeleton algorithm may have jagged edges, irregular branches, or unevenness, affecting navigation accuracy and visual effects. To ensure the geometric consistency and topological rationality of the optimized road network, the initial intra-layer road network can be further optimized to obtain the overall indoor road network.

[0059] For example, the initial intra-layer road network can be optimized using a multi-constraint overall indoor road network geometry optimization method to obtain the intra-layer road network. Further, inter-layer connectivity is performed on the intra-layer road networks of multiple layers to determine the overall indoor road network of the indoor space.

[0060] When optimizing the initial intra-layer road network, a topology graph can be constructed from the initial intra-layer road network through path simplification, adaptive resampling, and node vector calculation. Further constraint optimization of the topology graph based on constraints yields the intra-layer road network. For example, the original paths in the initial intra-layer road network can be geometrically simplified, and key geometric features can be extracted from the simplified original paths. Specifically, the Douglas-Peucker algorithm can be used to geometrically simplify the original paths in the obtained initial intra-layer road network, removing redundant points. For instance, paths smaller than 2 meters in the initial intra-layer road network can be simplified by merging endpoints to obtain the simplified original path.

[0061] Furthermore, the simplified original path can be sampled to obtain sampling nodes to ensure the preservation of key geometric features. For example, the type of the simplified original path can be determined first, and sampling can be implemented at different sampling intervals based on the type of the original path. The type of original path can include basic paths and fixed paths. A basic path can be a road, while a fixed path refers to a road connecting to a fixed point, such as a road connecting to an elevator, escalator, or door. The sampling interval can be determined according to actual needs; for example, the sampling interval for a normal path can be greater than the sampling interval for a fixed path.

[0062] Based on this, a topological relationship can be established between each sampling node in the initial layer road network, and the initial layer road network can be constructed into a topological graph according to the topological relationship. For example, a connection direction vector between sampling nodes can be calculated for each sampling node in the initial layer road network to establish a topological relationship between sampling nodes, and the initial layer road network can be constructed into a topological graph according to the topological relationship.

[0063] After obtaining the topology map, the road shapes in the topology map can be optimized. Specifically, spatial constraints and multiple constraint conditions can be constructed. Under these spatial constraints and multiple constraint conditions, the topology map is optimized through nonlinear least squares solution to determine the intra-layer road network. The spatial constraints can be determined based on the constructed distance field. Specifically, the spatial constraints are used to find the optimal path location, ensuring that the corrected path is located as close as possible to the center of vertical transportation facilities and away from obstacles and walls. In some embodiments, a first distance from the path location to the road area and a second distance from the path location to the obstacle area can be calculated. The first distance can be calculated based on a distance transformation; the larger the first distance, the safer the location and the further away from obstacles. The second distance can be calculated based on an inverse distance transformation; the second distance represents the distance of each pixel to the nearest vertical transportation facility. When inside an obstacle, the second distance represents the depth into the obstacle. The distance field can be displayed on the binary map; the distance of each pixel constitutes the distance field. Based on this distance field, the distance of a pixel to the nearest obstacle can be seen. In the distance field, the closer to the center, the higher its weight.

[0064] Multiple constraints can include anchor point constraints, adjacency constraints, straightness constraints, and node constraints. Anchor point constraints can fix the position of key nodes, keeping the basic framework of the road network unchanged; adjacency constraints can control the distance between adjacent points, maintaining path continuity; straightness constraints can constrain collinearity of three points, reducing unnecessary curvature, and using angles to control path smoothness; node constraints can standardize the geometry of multi-way intersection nodes, ensuring the rationality of intersection angles. After obtaining multiple constraints, the topology map can be optimized using nonlinear least-squares global optimization based on the Ceres library to obtain the optimized road network vector geometry, thereby determining the road network within each layer. Specifically, the above constraints are integrated into an optimization objective function, which is then solved using the Ceres library. During the optimization process, jagged edges and irregular branches are eliminated to improve geometric quality; road network connectivity is maintained to preserve topological rationality; and smoother, more realistic paths are provided to enhance navigation accuracy.

[0065] For the optimized intra-floor road network, the intra-floor road networks of each floor can be interconnected to obtain the overall indoor road network of the indoor space. For example, the traffic relationships of vertical transportation facilities are vectorized into roads, and the nodes of vertical transportation facilities are connected by edges to connect the intra-floor road networks, thus obtaining the overall indoor road network. Specifically, the traffic relationships of multiple vertical transportation facilities between different floors are obtained from the reference traffic relationships of elevators and / or escalators stored in the database. Further, the type of vertical transportation facility can be determined by the traffic relationships, and bidirectional or unidirectional traffic edges can be constructed between nodes according to the type of vertical transportation facility. For example, if the type of vertical transportation facility is determined to be an elevator based on the traffic relationships, bidirectional traffic edges are constructed between the nodes of multiple vertical transportation facilities; if the type of vertical transportation facility is determined to be an escalator based on the traffic relationships, unidirectional edges are constructed between the nodes of multiple vertical transportation facilities. In addition, the traffic relationships of vertical transportation facilities between different floors can be vectorized into roads. By connecting the nodes of vertical transportation facilities that have a passage relationship with each other through edges, the connectivity between the road networks within a floor is achieved, thus obtaining the overall indoor road network. For example, elevator nodes can be connected through bidirectional passage edges, and escalator nodes can be connected through unidirectional passage edges.

[0066] For example, parsing the escalator / elevator access relationships stored in the database, if elevator A on the 1st floor can connect to elevator B on the 2nd floor, the access relationship is vectorized into roads. The nodes corresponding to elevators A and B are connected by a bidirectional access edge, achieving connectivity between the intra-floor road networks and obtaining the overall indoor road network. It's important to note that when determining access relationships, access between elevators is bidirectional, but access between escalators is unidirectional.

[0067] Modeling of bidirectional elevator traffic, modeling of unidirectional escalator traffic, and cross-floor node connection mechanisms support cross-floor path planning, integrating the intra-floor road network into a unified three-dimensional network, and enabling the construction of a complete overall indoor road network.

[0068] It should be added that the generated overall indoor road network is in meter coordinate system. In order to improve accuracy, the overall indoor road network needs to be mapped back to latitude and longitude coordinates. At the same time, the road data and node data need to be standardized to achieve coordinate mapping and standardization processing.

[0069] Figure 4 The flowchart illustrating the generation of the overall indoor road network is shown in the figure. (Refer to...) Figure 4 As shown, the main steps include: Step S410 involves extracting the layered structure of the interior space to generate layered geometric data. This includes extracting layered structures, shop details, doors, elevators, escalators, etc.

[0070] Step S420, Image Feature Extraction. Binary image feature extraction is performed on the layered geometric data to obtain image features. Image features can be represented as a binary image, for example, as shown below. Figure 5 As shown in the image.

[0071] Step S430: Morphological road skeleton extraction is performed on the image features to obtain skeleton lines, thereby determining the basic road network.

[0072] Step S440: Add a reachable path for a fixed POI. A fixed POI can be a vertical transportation facility such as an elevator or escalator. Specifically, connect the vertical transportation facility to the basic road network to determine the connecting road network.

[0073] Step S450, Arterial Road Extraction. Arterial roads are extracted based on the combined road network consisting of connecting road networks and basic road networks to determine the initial intra-layer road network. Intra-layer road network refers to the road network of each layer.

[0074] Step S460, Road shape optimization. The initial intra-layer road network is optimized based on multiple constraints to obtain the intra-layer road network.

[0075] For example, refer to Figure 6 As shown, road skeletons can be extracted from layered geometric data to determine the basic road network of the road area. Vertical transportation facilities are then connected to the basic road network to determine the connecting road network. Based on the combined road network formed by the connecting road network and the basic road network, an initial intra-layer road network is determined, and this initial intra-layer road network is then optimized to obtain the final intra-layer road network.

[0076] Step S470: Import Inter-floor Access Relationships. Connect the intra-floor road network to different floors and import the inter-floor access relationships to obtain the overall indoor road network. (Reference) Figure 7 As shown, the overall indoor road network of the entire shopping mall is obtained by connecting the intra-floor road network of each floor with inter-floor connections.

[0077] Step S480, Specification Output. Map the overall indoor road network to latitude and longitude coordinates, standardize it, and add the overall indoor road network to the original overall indoor road network, making it part of the road network map.

[0078] In some embodiments, after obtaining the overall indoor road network, if it is detected that the difference result indicates that the current distribution data has changed or the data has been added compared to the historical distribution data, the obtained overall indoor road network can be added to the original overall indoor road network of the indoor space to update the original overall indoor road network of the indoor space.

[0079] For example, if the difference results indicate that some content in the overall indoor road network needs to be deleted, an update list for the indoor space can be determined. The update list can include deletion information and update objects represented by change information. Deletion information can include deleting roads, deleting nodes, etc. Change information can also include changing roads, changing nodes, etc. The identification information of the update objects in the update list is obtained from the first database. The overall indoor road network related to the identification information in the original overall indoor road network is deleted using the identification information, thereby updating the original overall indoor road network.

[0080] Based on the difference results, a list of deleted and changed items in the shopping mall is compiled. Both categories require the deletion of the existing online overall indoor road network of the shopping mall. Specifically, the indoor road and node data of the shopping mall to be deleted from the first database are read, and the original overall indoor road network of the indoor space is deleted using the identification information of the indoor road and node data.

[0081] In other embodiments, the outdoor road network can be acquired, and the connection relationship between the overall indoor road network and the outdoor road network can be determined based on the door location information of the indoor space, thereby realizing the connection between the overall indoor road network and the outdoor road network. Specifically, the overall indoor road network refers to the three-dimensional vector road network inside the indoor space, which is in an isolated state. The outdoor road network can be determined based on street view, trajectory, or multi-source data of the location of the indoor space. Furthermore, the connection relationship between the overall indoor road network and the outdoor road network can be determined based on the door location information of the indoor space.

[0082] After updating the original overall indoor space based on the overall indoor road network and determining the connection relationship between the overall indoor road network and the outdoor road network, the online map road network data related to the indoor space can be updated. Online map road network data refers to detailed map data associated with the indoor space, which can be used for applications such as navigation.

[0083] Figure 8 The diagram illustrates a flowchart of the automated update process for the overall indoor road network of the shopping mall. (Reference) Figure 8 As shown, the main steps include: In step S810, the server acquires and preprocesses data from the first database, such as acquiring current and historical distribution data of the indoor space.

[0084] Step S820, Data Differentiation. Specifically, the difference between the latest data in the first database and the online data is calculated in the server. For example, the difference between the current distribution data and the historical distribution data is used as the difference result.

[0085] Step S830: The server generates the overall indoor road network.

[0086] For example, if the server determines that there is new data based on the difference results, a new overall indoor road network is automatically generated according to steps S110 to S130; if the difference results determine that there is data deletion, the algorithm automatically generates information on deleting the overall indoor road network of the online shopping mall and executes the deletion of the overall indoor road network; if the server determines that there is data change based on the difference results, the algorithm automatically generates information on deleting the old online shopping mall and generates the latest overall indoor road network, thereby updating the original overall indoor road network.

[0087] In step S840, the server updates the online map road network data in the second database based on the generated overall indoor road network.

[0088] The outdoor road network is connected using the gate location information. A connection relationship between the overall indoor and outdoor road networks is established based on the gate location information to link the two networks.

[0089] After updating the overall indoor road network and connecting it with the outdoor road network, the online map road network data in the second database is updated. This allows one or more clients to read the updated online map road network data from the second database and navigate based on it. The second database can be a database used for online display.

[0090] Step S850: Display the updated online map road network data to the client.

[0091] In this embodiment of the disclosure, the update of the overall indoor road network and the update of the online map road network data are only required when it is determined, based on the difference results of the current distribution data and the historical distribution data, that the overall indoor road network needs to be generated or the overall indoor road network needs to be deleted. This avoids the problem of large workload caused by frequent updates of online data and reduces resource consumption.

[0092] The technical solution in this disclosure automates the entire process from data update to road network generation. Furthermore, by incrementally updating the overall indoor road network, it avoids full reconstruction and improves operational efficiency. Optimization through multiple constraints ensures the overall quality of the indoor road network while maintaining topological rationality.

[0093] The complex 3D problem is decomposed into 2D structures and connectivity relationships, and the operation process is simplified through hierarchical processing. Road network quality is gradually improved through multiple iterations, and multiple constraints guide path correction, thereby enhancing the overall accuracy and quality of the indoor road network and improving the efficiency and quality of indoor map production.

[0094] This disclosure also provides a road network update system, applicable to the updating process of online map road network data for indoor spaces. (See reference...) Figure 9As shown, the road network update system 900 may include: The first database, 901, is used to store the current and historical distribution data of indoor spaces. Server 902, connected to the first database, is used to generate an overall indoor road network when it is determined that an overall indoor road network needs to be generated based on the difference results between the current distribution data and the historical distribution data, and to update the online map road network data of the indoor space based on the overall indoor road network; the overall indoor road network is obtained according to the above road network generation method; The second database, 903, is used to store updated online map road network data; Client 904 is used to retrieve updated online map road network data from the second database for display.

[0095] In this embodiment, the first database can be any type of database used for storing data, and the second database can be a database used for online display. The client can be a mobile phone, tablet computer, personal computer, smart TV, smartwatch, smart speaker, smart vehicle device, and wearable device, but is not limited to these.

[0096] A server can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server or cloud server cluster that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, and big data and artificial intelligence platforms, but it is not limited to these.

[0097] The server also includes: a connection relationship determination module, used to acquire the outdoor road network and determine the connection relationship between the overall indoor road network and the outdoor road network based on the door location information of the indoor space; and an update control module, used to update the original overall indoor space of the indoor space according to the overall indoor road network and, after determining the connection relationship, update the online map road network data of the indoor space in the second database according to the overall indoor road network.

[0098] The update control module includes: a difference result determination module, used to obtain the difference result between the current distribution data and historical distribution data of the indoor space; a generation module, used to determine the need to generate an overall indoor road network based on the difference result, and update the original overall indoor road network of the indoor space based on the overall indoor road network; and a deletion module, used to determine the need to delete the overall indoor road network based on the difference result, determine the update list of the indoor space, obtain the identification information of the update objects in the update list from the database, and perform a deletion operation on the original overall indoor road network of the indoor space through the identification information to update the overall indoor road network.

[0099] In this embodiment of the disclosure, the online map road network data needs to be updated only when it is determined, based on the difference results of the current distribution data and the historical distribution data, that it is necessary to generate or delete the overall indoor road network. This avoids the problem of large workload caused by frequent updates of online data and reduces resource consumption.

[0100] It should be noted that the specific details of each part of the above-mentioned road network update system have been described in detail in some implementations of the corresponding methods. For details that are not disclosed, please refer to the implementation content of the method section, and therefore will not be repeated here.

[0101] Exemplary embodiments of this disclosure also provide an electronic device. This electronic device can be the aforementioned terminal device or server. Generally, the electronic device may include a processor and a memory, the memory storing executable instructions of the processor, and the processor configured to execute the aforementioned road network generation method by executing the executable instructions. Furthermore, the electronic device may also include a display for displaying an interface. The electronic device can be a smartphone or a vehicle capable of executing the aforementioned road network generation method; any device capable of implementing the road network generation method is acceptable, and no specific limitation is made here.

[0102] The electronic device is described below as an example in the form of a general-purpose computing device. This electronic device is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0103] The components of an electronic device may include, but are not limited to: at least one processing unit, at least one storage unit, a bus connecting different system components (including storage units and processing units), and a display unit.

[0104] The storage unit stores program code that can be executed by the processing unit, causing the processing unit to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure. For example, the processing unit can perform actions such as... Figure 1 The steps are shown in the figure.

[0105] The storage unit may include readable media in the form of volatile storage units, such as random access memory (RAM) and / or cache storage units, and may further include read-only memory (ROM).

[0106] The storage unit may also include a program / utility having a set (at least one) of program modules, including but not limited to: an operating system, one or more applications, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0107] A bus can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus that uses any of the various bus structures.

[0108] The electronic device can also communicate with one or more external devices (such as keyboards, pointing devices, Bluetooth devices, etc.), one or more devices that enable a user to interact with the electronic device, and / or any device that enables the electronic device to communicate with one or more other computing devices (such as routers, modems, etc.). This communication can be performed via input / output (I / O) interfaces. Furthermore, the electronic device can communicate with one or more networks (such as local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter. As shown in the figure, the network adapter communicates with other modules of the electronic device via a bus. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0109] It should be noted that some embodiments of this disclosure also provide a computer program product, which includes a computer program that implements the above-described method when executed by a processor.

[0110] In one implementation, the computer program product can be a tangible product containing a computer program, such as a computer-readable storage medium storing the computer program. The readable storage medium can be a storage medium based on electrical, magnetic, optical, electromagnetic, infrared, or other signals, including but not limited to: random access memory (RAM), read-only memory (ROM), magnetic tape, floppy disk, flash memory, hard disk drive (HDD), solid-state drive (SSD), etc. For example, the computer program product can be implemented as a non-volatile storage medium storing a computer program, such as read-only memory, NAND flash memory, etc.

[0111] In one implementation, the computer program product can be an intangible product containing a computer program. For example, the computer program product can be implemented as a virtual digital product, such as an executable file, installation package, or other digital file storing the computer program.

[0112] Computer program code can be written in one or more programming languages. Examples of programming languages ​​include C, Java, and C++. Program code can execute entirely on the user's computing device, partially on the user's computing device, or as a standalone software package. It can also execute partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, such as a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via an internet connection provided by a mobile network operator).

[0113] Computer programs can be carried or transmitted via signals such as electrical, magnetic, optical, electromagnetic, and infrared rays. Electronic devices can convert signals carrying computer programs into digital signals, thereby running the computer programs. When a computer program runs on an electronic device, its code is used to cause the electronic device to execute (more specifically, to be executed by the processor of the electronic device) the method steps of various exemplary embodiments of this disclosure.

[0114] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0115] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0116] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A road network generation method, characterized in that, include: Obtain layered geometric data of the indoor space, and construct a graph structure of the road area in the indoor space based on the layered geometric data to determine the basic road network; Nodes of vertical transportation facilities are identified and connecting roads between the nodes and the basic road network are generated to construct the connecting road network between the vertical transportation facilities and the basic road network. The main roads of the combined road network consisting of the basic road network and the connecting road network are extracted to determine the initial intra-layer road network. The initial intra-layer road network is optimized to obtain the intra-layer road network. The intra-layer road network is then connected between layers to obtain the overall indoor road network.

2. The road network generation method according to claim 1, characterized in that, The step of constructing a graph structure of road areas in the indoor space based on the layered geometric data to determine the basic road network includes: Binary image feature extraction is performed on the layered geometric data to determine image features; Based on the image features, morphological skeleton extraction is performed, refining the road region in the image features into a center line with a single pixel width, while maintaining the topological connectivity of the road region, in order to determine the skeleton line; Key nodes are identified on the skeleton line, and the pixel sequence contained in each edge of the path segment connecting each key node is extracted. A graph structure containing the node coordinates and road coordinates of the key nodes is constructed to determine the basic road network.

3. The road network generation method according to claim 1, characterized in that, The process of determining nodes for each vertical transportation facility and generating a connection network between those nodes and the basic road network includes: Using the center point of the vertical transportation facility as a node, and taking the node as the starting point, search for the road closest to the starting point; Based on the road, determine the connecting road between the node and the basic road network, and obtain the connecting road network based on the connecting road.

4. The road network generation method according to claim 1, characterized in that, The step of extracting arterial roads to determine the initial intra-layer road network from the combined road network formed by the basic road network and the connecting road network includes: Convert the combined road network into an undirected graph; The undirected graph contains a loop structure, and adding a point will interrupt the loop structure; If there are edges of the first type in the undirected graph after the iterative query is interrupted, delete the edges of the first type until there are no edges of the first type in the undirected graph, so as to determine the first undirected graph; For each edge in the first undirected graph, determine the width data, iteratively determine the existence of edges of the second type based on the width data, delete the edges of the second type, until there are no edges of the second type in the undirected graph, so as to obtain the second undirected graph; The edges connected to nodes of degree 2 in the second undirected graph are merged to generate the initial intra-layer road network.

5. The road network generation method according to claim 1, characterized in that, The optimization of the initial intra-layer road network to obtain the intra-layer road network includes: The original paths in the road network within the initial layer are geometrically simplified, and the simplified original paths are sampled to determine the sampling nodes; Establish topological relationships between nodes for each sampling node in the initial layer road network, and construct a topological graph based on the topological relationships of the initial layer road network; A distance field is constructed to determine spatial constraints and multiple constraint conditions. Based on the spatial constraints and multiple constraint conditions, the topology graph is optimized by nonlinear least squares solution to determine the intra-layer road network.

6. The road network generation method according to claim 1, characterized in that, The step of determining the overall indoor road network by performing inter-layer connectivity on the intra-layer road network includes: The passage relationships between multiple vertical transportation facilities at different levels are parsed from the reference passage relationships stored in the first database; Based on the traffic relationship, the type of the vertical transportation facility is determined to be an elevator, and a two-way traffic edge is constructed between the nodes of multiple vertical transportation facilities; Based on the traffic relationship, the type of the vertical transportation facility is determined to be an escalator, and a one-way traffic edge is constructed between the nodes of multiple vertical transportation facilities; The traffic relationships of the vertical transportation facilities are vectorized into roads, and the nodes of multiple vertical transportation facilities are connected through bidirectional or unidirectional traffic edges to obtain the overall indoor road network.

7. A road network renewal system, characterized in that, include: The first database is used to store the current and historical distribution data of indoor spaces; The server, communicatively connected to the first database, is used to generate an overall indoor road network when it is determined that an overall indoor road network needs to be generated based on the difference result between the current distribution data and the historical distribution data, and to update the online map road network data of the indoor space based on the overall indoor road network; the overall indoor road network is obtained by the road network generation method according to any one of claims 1-6; The second database is used to store updated online map road network data; The client is used to retrieve updated online map road network data from the second database for display.

8. The road network renewal system according to claim 7, characterized in that, The server also includes: The connection relationship determination module is used to acquire the outdoor road network and determine the connection relationship between the overall indoor road network and the outdoor road network based on the door location information of the indoor space. The update control module is used to update the original overall indoor space of the indoor space according to the overall indoor road network, and after determining the connection relationship, update the online map road network data of the indoor space in the second database according to the overall indoor road network.

9. The road network generation method according to claim 8, characterized in that, The update control module includes: The difference result determination module is used to obtain the difference result between the current distribution data and the historical distribution data of the indoor space; The generation module is used to determine the overall indoor road network that needs to be generated based on the difference results, and to update the original overall indoor road network of the indoor space based on the overall indoor road network. The deletion module is used to determine the overall indoor road network that needs to be deleted based on the difference results, determine the update list of the indoor space, obtain the identification information of the update objects in the update list from the database, and perform a deletion operation on the original overall indoor road network of the indoor space through the identification information to update the overall indoor road network.

10. A computer program product, characterized in that, When the computer program is executed by the processor, it implements the road network generation method according to any one of claims 1-6.

Images