Three-dimensional road network extraction method and device based on engineering model and vehicle positioning information

Abstract

The application discloses a three-dimensional road network extraction method and equipment based on an engineering model and vehicle positioning information, and relates to the field of three-dimensional geographic information processing.The method comprises the following steps: acquiring vehicle positioning data of a target mountainous area; projecting a BIM model according to the plane coordinates of the vehicle positioning data of any first road point to obtain a plurality of BIM projection points; projecting a GIS real scene model according to the plane coordinates of the vehicle positioning data of any second road point to obtain a plurality of GIS projection points; the first road point is any positioning point of the vehicle positioning data collected by a positioning module in an underground tunnel; the second road point is any positioning point of the vehicle positioning data of any road section; selecting the three-dimensional coordinates of the BIM projection point with the lowest Z coordinate as the three-dimensional coordinates of the any first road point, and selecting the three-dimensional coordinates of the GIS projection point with the lowest Z coordinate as the three-dimensional coordinates of the any second road point; and the application can accurately and comprehensively acquire three-dimensional road data.

Description

Technical Field

[0001] This application relates to the field of three-dimensional geographic information processing technology, and in particular to a method and device for extracting three-dimensional road networks based on engineering models and vehicle positioning information. Background Technology

[0002] In the process of developing intelligent hydropower and digital dam construction, the three-dimensional data of mountain roads around the dam is the core foundation for intelligent dam operation and maintenance, construction route planning and surrounding three-dimensional modeling. Its accuracy and completeness directly determine the quality of related engineering applications and technology research and development.

[0003] Currently, the main methods for acquiring 3D road data in the mountainous areas surrounding the dam include GPS positioning combined with elevation fitting, and directly extracting 3D road data based on BIM models or CAD road network data. However, the positioning accuracy of GPS positioning combined with elevation fitting is insufficient, especially the elevation positioning accuracy error, which is usually more than 5 meters, far lower than the planar accuracy. Furthermore, GPS signal loss occurs in environments such as underground tunnels, making it impossible to obtain 3D road data for underground caverns and other similar environments, resulting in incomplete 3D road data. The update mechanisms of BIM models and CAD road network data are cumbersome and have a low update frequency, resulting in poor real-time performance and incomplete data for 3D road data extracted from BIM models and CAD road network data, thus leading to low data accuracy. In conclusion, existing 3D road data extraction methods cannot accurately and comprehensively acquire 3D road data in the mountainous areas surrounding the dam. Summary of the Invention

[0004] The purpose of this application is to provide a method and device for extracting three-dimensional road networks based on engineering models and vehicle positioning information, which can accurately and comprehensively acquire three-dimensional road data of the mountainous area surrounding the dam.

[0005] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for extracting a three-dimensional road network based on an engineering model and vehicle positioning information, including: Obtain vehicle location data for the target mountainous area; Based on the planar coordinates of the vehicle positioning data of any first road point, a BIM model is projected to obtain several BIM projection points. Based on the planar coordinates of the vehicle positioning data of any second road point, a GIS real-scene model is projected to obtain several GIS projection points. The first road point is any positioning point in the vehicle positioning data of any underground tunnel road, and the vehicle positioning data of any underground tunnel road is collected by a positioning module in the underground tunnel. The second road point is any positioning point in the vehicle positioning data of any section of any non-underground tunnel road. The 3D coordinates of the BIM projection point with the lowest Z-coordinate are selected as the 3D coordinates of any first road point, and the 3D coordinates of the GIS projection point with the lowest Z-coordinate are selected as the 3D coordinates of any second road point, thus obtaining the 3D road network data of the target mountain area.

[0006] Optionally, the three-dimensional road network extraction method based on engineering models and vehicle positioning information acquires CAD road network data of the target mountain area and gate information of various underground tunnel entrances while acquiring vehicle positioning data of the target mountain area; wherein, the CAD road network data includes road coordinates and road attributes, the road attributes include underground tunnels and non-underground tunnels, and the gate information includes at least the number of the underground tunnel. The method for extracting a 3D road network based on an engineering model and vehicle positioning information further includes, before projecting the BIM model based on the planar coordinates of the vehicle positioning data of any first road point: Based on the road attributes of the target mountainous area and the gate information of each underground tunnel entrance, the system detects whether any road point has corresponding CAD road network data; wherein, the road point includes the first road point and the second road point; The process of projecting a BIM model based on the planar coordinates of vehicle positioning data from any first road point to obtain several BIM projection points, and projecting a GIS reality model based on the planar coordinates of vehicle positioning data from any second road point to obtain several GIS projection points, specifically includes: If any of the road points has corresponding CAD road network data, then the vertical projection correction operation and the engineering model projection operation are executed sequentially. If no corresponding CAD road network data exists for any of the road points, perform an engineering model projection operation; The vertical projection correction operation includes: Based on the CAD road network data of any road point, the planar coordinates of the vehicle positioning data of any road point are determined. x , y Perform planar distance threshold correction to obtain the corrected coordinates of any road point; The engineering model projection operation includes: Projecting the target coordinates of any first road point along the Z-axis onto the BIM model of the corresponding underground tunnel yields several BIM projection points; and / or projecting the target coordinates of any second road point along the Z-axis onto the GIS real-world model of the target mountain area yields several GIS projection points; wherein the target coordinates are calibrated coordinates or planar coordinates of vehicle positioning data.

[0007] Optionally, the planar coordinates of the vehicle positioning data of any road point are based on the CAD road network data of any road point.x , y Perform planar distance threshold correction to obtain the corrected coordinates of any road point, specifically including: Calculate the planar coordinates of the vehicle positioning data for any of the road points. x , y Find the shortest distance to the corresponding CAD road network data segment, and detect whether the shortest distance is less than a preset distance threshold. If so, project any road point onto the CAD road network data segment along a direction perpendicular to the CAD road network data segment to obtain the corrected coordinates of any road point. Projecting the target coordinates of any first road point along the Z-axis onto the BIM model of any underground tunnel yields several BIM projection points, specifically including: Based on the number of any underground tunnel and the component attributes of the BIM model of the target mountain area, the BIM component of any underground tunnel is selected from the BIM components of the BIM model; wherein, by parsing the BIM model of the target mountain area, each BIM component and component attributes of the BIM model are obtained. Project the target coordinates of any first road point along the negative Z-axis onto the BIM component of any underground tunnel to obtain several BIM projection points. Projecting the target coordinates of any second road point along the Z-axis onto the GIS real-world model of the target mountain area yields several GIS projection points, specifically including: Based on the quadtree spatial index structure, according to the target coordinates of any second road point, the set of triangular faces within the spatial region where the second road point is located is retrieved as a subset of candidate triangular faces; wherein, the quadtree spatial index structure is used to parse the GIS real-world model to obtain a set of triangular face meshes of the GIS real-world model, and based on the planar coordinates of the vertices of the triangular faces of the GIS real-world model ( x , y The triangular mesh quadtree spatial index tree obtained by constructing the triangular mesh quadtree spatial index tree; Project the target coordinates of any second road point along the negative Z-axis onto the candidate triangular face subset to obtain several GIS projection points.

[0008] Optionally, the three-dimensional road network extraction method based on engineering models and vehicle positioning information, before detecting whether corresponding CAD road network data exists for any road point based on the road attributes of the target mountainous area and the gate information of various underground tunnel entrances, further includes: The acquired vehicle positioning data and coordinates of the GIS real-scene model of the target mountain area are uniformly converted to the right-hand rectangular two-dimensional coordinate system of the BIM model of the target mountain area, and the converted two-dimensional coordinates of the vehicle positioning data of the target mountain area are then converted. x , y The planar coordinates of the vehicle positioning data are used as the two-dimensional coordinates after the GIS real-scene model of the target mountain area is converted. x , y () serves as the planar coordinates of the GIS real-world model.

[0009] Optionally, the three-dimensional road network extraction method based on engineering models and vehicle positioning information, before detecting whether corresponding CAD road network data exists for any road point based on the road attributes of the target mountainous area and the gate information of various underground tunnel entrances, further includes: Based on the gate information of various underground tunnel entrances in the target mountainous area, the time range of the target vehicle being located in each underground tunnel is obtained; wherein, the gate information also includes the time of the target vehicle entering and exiting each underground tunnel; Based on the time range, the vehicle positioning data of the target vehicle in the target mountain area is screened to obtain the vehicle positioning data of the target vehicle in various underground tunnels in the target mountain area and the vehicle positioning data of the target vehicle on non-underground tunnel roads in the target mountain area.

[0010] Optionally, the method for extracting a 3D road network based on an engineering model and vehicle positioning information further includes: Calculate the normal vector of the triangle face containing the BIM projection point with the lowest Z coordinate; Determine the angle between the normal vector and the vertical direction. ; Detect the included angle If the angle is less than a preset threshold, and if it is not less than a preset threshold, then any of the first road points is determined to be the first abnormal road point and is removed.

[0011] Optionally, the method for extracting a 3D road network based on an engineering model and vehicle positioning information further includes: Sort the three-dimensional coordinates of all road points according to their timestamps to obtain the three-dimensional data of the first target. The second abnormal road points in the first target's three-dimensional data are removed to obtain the third target's three-dimensional data, and the missing road point data in the third target's three-dimensional data are filled in by linear interpolation; wherein, the second abnormal road points include road points with abnormal speed and road points with abnormal slope.

[0012] Optionally, removing the second abnormal road points from the three-dimensional data of the first target specifically includes: Based on the timestamp of each road point in the first target 3D data, each road point in the first target 3D data is sequentially treated as a first target point and speed processing is performed until the second to last road point in the first target 3D data is treated as a first target point and the speed processing operation is completed, thus obtaining the second target 3D data. According to the timestamp of each road point in the second target 3D data, the slope processing operation is performed on each road point in the second target 3D data as the second target point in turn, until the slope processing operation is performed on the second to last road point in the second target 3D data as the second target point, and the third target 3D data is obtained. The speed processing operation includes: Calculate the Euclidean distance between the first target point and the adjacent road points; The vehicle speed is obtained by dividing the Euclidean distance by the sampling time interval between the first target point and the adjacent road point; If the vehicle's moving speed is greater than a set speed threshold, then the adjacent road points of the first target point are considered to be speed anomaly points and are removed. The slope processing operation includes: Calculate the horizontal distance between the second target point and the adjacent road point; Based on the horizontal distance, the Z-coordinate of the second target point and the Z-coordinate of the adjacent road point of the second target point, calculate the road slope between the second target point and the adjacent road point; If the road slope is greater than a set slope threshold, the adjacent road points of the second target point are considered to be slope anomaly points and are removed.

[0013] Optionally, the three-dimensional road network extraction method based on engineering models and vehicle positioning information, after removing the second abnormal road points from the first target three-dimensional data and filling in the removed second abnormal road points using linear interpolation, further includes: The three-dimensional data of the third target is discretized at the meter level to obtain discretized road points, and a w×w two-dimensional grid road covering all valid road points in the three-dimensional data of the third target is constructed, where w is the width; Map all discrete road points to the corresponding grids of the two-dimensional grid roads, and count the number of roads covered by the discrete road points in each grid of the two-dimensional grid roads to obtain the number of roads covered by each grid. Check if the number of roads covered by each grid is less than 2; If any grid cell covers a road segment with a number of roads less than 2 and a number of roads equal to 1, then the road segment is determined to be a regular road segment. If the number of roads covered by any grid is not less than 2, then it is detected whether there are at least two roads in the road covered by the grid that have an intersection area and the elevation difference is less than the set first elevation difference threshold. If so, it is considered that there is an intersection segment in the grid and the grid is determined to be an intersection area. Extract the valid road points of the intersection segments in each intersection area, and calculate the geometric center point of the extracted valid road points as the intersection breakpoint; The intersection area is split from each intersection breakpoint, and redundant and overlapping road points at each intersection breakpoint are removed to obtain independent road segments that do not intersect and are connected only at the intersection breakpoints. The planar coordinates of the valid road points of the ordinary road segments and independent road segments are transformed to the WGS-84 coordinate system and / or CGCS2000 coordinate system, and then encapsulated as GeoJSON, SHP, DXF or 3D model format for output.

[0014] Secondly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the three-dimensional road network extraction method based on engineering models and vehicle positioning information as described in any of the above claims.

[0015] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a method and device for extracting a 3D road network based on an engineering model and vehicle positioning information. By acquiring vehicle positioning data of a target mountainous area, and using the planar coordinates of the vehicle positioning data (collected by a positioning module inside an underground tunnel) of any first road point (any positioning point in the vehicle positioning data of any underground tunnel road), a BIM model is projected to obtain several BIM projection points. Then, using the planar coordinates of the vehicle positioning data of any second road point, a GIS real-scene model is projected to obtain several GIS projection points. The 3D coordinates of the BIM projection point with the lowest Z-coordinate are selected as the 3D coordinates of the first road point, and the 3D coordinates of the GIS projection point with the lowest Z-coordinate are selected as the 3D coordinates of the second road point. This yields the 3D road network data of the target mountainous area, enabling comprehensive extraction of 3D road network data for both underground and non-underground tunnel roads in the target mountainous area. This solves the problem of GPS signal loss in environments such as underground tunnels, where existing GPS positioning combined with elevation fitting methods cannot obtain 3D road data for underground caverns and other environments, resulting in incomplete 3D road data.

[0016] Meanwhile, by combining vehicle-mounted positioning data with BIM and GIS reality models, the obtained elevation coordinates of underground and non-underground tunnel roads are more accurate, thus solving the problem of insufficient elevation positioning accuracy caused by GPS positioning combined with elevation fitting. Furthermore, compared to existing methods that extract 3D road data solely based on BIM models, the use of GIS reality models avoids the low accuracy problem caused by poor real-time performance in BIM model-based 3D road data extraction, resulting in better real-time performance and thus more accurate data. In addition, the 3D road network data obtained through vehicle-mounted positioning data has better real-time performance and is more comprehensive and accurate than traditional CAD road network data, solving the problems of poor real-time performance and low accuracy in 3D road data extracted from CAD road network data. Therefore, the embodiments of this application can accurately obtain 3D road data of the mountainous area surrounding the dam.

[0017] In summary, the three-dimensional road data extraction method of this application embodiment can accurately and comprehensively obtain three-dimensional road data of the mountainous area surrounding the dam. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating a three-dimensional road network extraction method based on an engineering model and vehicle positioning information, provided in an embodiment of this application; Figure 2 A structural block diagram of a three-dimensional road network extraction system based on an engineering model and vehicle positioning information provided in one embodiment of this application; Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] In one exemplary embodiment, such as Figure 1 As shown, a method for extracting a 3D road network based on an engineering model and vehicle positioning information is provided. This method is executed by a computer device and includes the following steps 101 to 103. Wherein: Step 101: Obtain vehicle positioning data for the target mountainous area.

[0023] In this embodiment of the application, the target mountainous area can be the mountainous area surrounding the target dam. The specific range of the mountainous area surrounding the target dam can be determined according to actual needs, and is not specifically limited here.

[0024] The vehicle positioning data for each location point is arranged in order of timestamp, and the data structure is similar (time, longitude, latitude, elevation (inaccurate)).

[0025] Step 102: Based on the planar coordinates of the vehicle positioning data of any first road point, perform BIM model projection to obtain several BIM projection points, and based on the planar coordinates of the vehicle positioning data of any second road point, perform GIS real-scene model projection to obtain several GIS projection points; wherein, the first road point is any positioning point in the vehicle positioning data of any underground tunnel road, and the vehicle positioning data of any underground tunnel road is collected by the positioning module in the underground tunnel; the second road point is any positioning point in the vehicle positioning data of any section of any non-underground tunnel road.

[0026] Step 103: Select the 3D coordinates of the BIM projection point with the lowest Z coordinate as the 3D coordinates of any of the first road points, and select the 3D coordinates of the GIS projection point with the lowest Z coordinate as the 3D coordinates of any of the second road points, to obtain the 3D road network data of the target mountain area.

[0027] By implementing steps 101 to 103 above, vehicle positioning data of the target mountain area is acquired. Based on the planar coordinates of the vehicle positioning data (collected by the positioning module in the underground tunnel) of any first road point (any positioning point in the vehicle positioning data of any underground tunnel road), BIM model projection is performed to obtain several BIM projection points. Based on the planar coordinates of the vehicle positioning data of any second road point, GIS real-scene model projection is performed to obtain several GIS projection points. The three-dimensional coordinates of the BIM projection point with the lowest Z coordinate are selected as the three-dimensional coordinates of the first road point, and the three-dimensional coordinates of the GIS projection point with the lowest Z coordinate are selected as the three-dimensional coordinates of the second road point. The three-dimensional road network data of the target mountain area is obtained, realizing the comprehensive extraction of three-dimensional road network data of underground tunnel roads and non-underground tunnel roads in the target mountain area. This solves the problem that the existing GPS positioning combined with elevation fitting method has the problem of GPS signal loss in environments such as underground tunnels, thus failing to obtain three-dimensional road data of underground caverns and other environments, resulting in incomplete three-dimensional road data.

[0028] Meanwhile, by combining vehicle-mounted positioning data with BIM and GIS reality models, the obtained elevation coordinates of underground and non-underground tunnel roads are more accurate, thus solving the problem of insufficient elevation positioning accuracy caused by GPS positioning combined with elevation fitting. Furthermore, compared to existing methods that extract 3D road data solely based on BIM models, the use of GIS reality models avoids the low accuracy problem caused by poor real-time performance in BIM model-based 3D road data extraction, resulting in better real-time performance and thus more accurate data. In addition, the 3D road network data obtained through vehicle-mounted positioning data has better real-time performance and is more comprehensive and accurate than traditional CAD road network data, solving the problems of poor real-time performance and low accuracy in 3D road data extracted from CAD road network data. Therefore, the embodiments of this application can accurately obtain 3D road data of the mountainous area surrounding the dam.

[0029] In summary, the three-dimensional road data extraction method of this application embodiment can accurately and comprehensively obtain three-dimensional road data of the mountainous area surrounding the dam.

[0030] In another exemplary embodiment of this application, step 101 above acquires CAD road network data of the target mountain area and gate information of various underground tunnel entrances while acquiring vehicle positioning data of the target mountain area; wherein, the CAD road network data includes road coordinates and road attributes, the road attributes include underground tunnels and non-underground tunnels, and the gate information includes at least the number of the underground tunnel.

[0031] In this embodiment, CAD road network data is extracted from CAD road network drawings, providing precise two-dimensional coordinates of the road's centerline. While accurate, CAD road network data is limited and updates slowly, resulting in incomplete data. Road attributes include underground tunnels and non-underground tunnels, allowing for the filtering of non-underground tunnel roads and underground tunnel (cave) roads. When correcting vehicle positioning data, corresponding CAD road network data can be quickly matched based on road attributes. For example, for mountainous GPS data, CAD road network data for non-underground cavern roads can be filtered for correction; similarly, for vehicle positioning data of underground tunnel roads, CAD road network data with corresponding attributes can be filtered for correction.

[0032] The above-mentioned 3D road network extraction method based on engineering models and vehicle positioning information, before step 102, also includes: Step 201: Based on the road attributes of the target mountainous area and the gate information of each underground tunnel entrance, detect whether any road point has corresponding CAD road network data (including at least two-dimensional coordinate data); wherein, the road point includes the first road point and the second road point.

[0033] In the embodiments of this application, The gate information includes the number of the underground tunnel. This number can be used to filter out CAD road network data that has the attribute of underground tunnel and whose number matches the underground tunnel number in the gate information.

[0034] For vehicles on non-underground tunnel roads, GPS can be used to obtain vehicle positioning data. For vehicles in underground tunnels, UWB ultra-wideband positioning modules are preferred to obtain accurate positioning data.

[0035] Accordingly, step 102 specifically includes: Step 202: If any of the above road points has corresponding CAD road network data, perform vertical projection correction operation and engineering model projection operation in sequence.

[0036] The vertical projection correction operation includes: Based on the CAD road network data of any of the above road points, the planar coordinates of the vehicle positioning data of any of the above road points are ( x , y Perform planar distance threshold correction to obtain the corrected coordinates of any road point.

[0037] Step 203: If no corresponding CAD road network data exists for any of the above road points, perform the engineering model projection operation.

[0038] The engineering model projection operation includes: Projecting the target coordinates of any first road point along the Z-axis onto the BIM model of the corresponding underground tunnel yields several BIM projection points; and / or projecting the target coordinates of any second road point along the Z-axis onto the GIS real-world model of the target mountain area yields several GIS projection points; wherein the target coordinates are calibrated coordinates or planar coordinates of vehicle positioning data.

[0039] CAD road network data is generally considered accurate. For any road segment, if both vehicle positioning data and CAD road network data are available, the CAD road network data should be used as the standard, and the planar coordinates of the vehicle positioning data should be compared. x , y (The correction is performed.)

[0040] Here, projecting the target coordinates of any first road point along the Z-axis onto the BIM model of any of the aforementioned underground tunnels means that for any first road point... target coordinates Based on the maximum Z-value of all triangular faces in the BIM model, construct the corresponding 3D points. ( (Z), use a little more ( ,Z) Construct a ray along the negative Z-axis, and perform intersection detection between the constructed ray and the triangular mesh model of the corresponding underground tunnel BIM component to obtain several intersection points and the corresponding triangular face set. Each intersection point is a BIM projection point.

[0041] BIM models, through parametric modeling, construct digital models with precise geometric dimensions, structural attributes, engineering parameters, and spatial topological relationships. They can fully carry the design information, component relationships, and construction logic of engineering structures, and accurately reflect the three-dimensional form and engineering characteristics of engineering entities.

[0042] GIS reality models are generated based on real geographic space data. They are typically generated through multi-view images, laser point clouds, oblique photography, and other methods. They are expressed by combining triangular meshes with real-world textures and can achieve realistic, comprehensive, and high-precision replication of real geographic entities such as land surfaces, buildings, roads, vegetation, and water bodies. They include accurate spatial coordinates, geometric shapes, surface textures, geographic attributes, and other real-world scene information.

[0043] In this embodiment, the vehicle positioning data is corrected using CAD road network data to ensure the quality and reliability of the extracted three-dimensional road network data; by combining the corrected coordinates with BIM models and GIS real-scene models, the elevation coordinates of underground tunnel roads and non-underground tunnel roads are obtained more accurately.

[0044] In another exemplary embodiment of this application, in order to accurately obtain the correction coordinates, the planar coordinates of the vehicle positioning data of any of the aforementioned road points are obtained based on the CAD road network data of any of the aforementioned road points. x , y Perform planar distance threshold correction to obtain the corrected coordinates of any road point, specifically including: Step 301, calculate the planar coordinates of the vehicle positioning data for any of the above road points ( x , y Find the shortest distance to the corresponding CAD road network data segment and check if the shortest distance is less than a preset distance threshold. If so, project any road point onto the CAD road network data segment along a direction perpendicular to the CAD road network data segment to obtain the corrected coordinates of any road point.

[0045] In this embodiment of the application, the corresponding CAD road network data refers to the CAD road network data of any road point.

[0046] For any road point in a two-dimensional plane, calculate the shortest distance from it to the road segment in the CAD road network data, that is, calculate the shortest distance from it to each of the constituent line segments in the road segment composed of multiple line segments. Specifically, for any line segment, determine the projection point of the target point on the line segment using the vector projection method. When the projection point is inside the line segment, calculate the perpendicular distance from any road point to the line segment, which is taken as the distance from any road point to the line segment. When the projection point is outside the two endpoints of the line segment, calculate the Euclidean distance from any road point to the nearest endpoint, which is taken as the distance from any road point to the line segment. The nearest endpoint is the endpoint of the line segment that is closest to any road point. Finally, take the minimum value among the distances from any road point to all line segments as the shortest distance from any road point to the multiple line segments.

[0047] CAD road network data is generally considered accurate. For any road segment, when both vehicle positioning data and CAD road network data are available, the CAD road network data is used as the standard. The criterion for judgment is whether the distance between the vehicle positioning data and the CAD road network data is less than the distance threshold D1. If it is less, it is considered that the vehicle positioning data and the CAD road network data describe the same road. Therefore, the vehicle positioning data is corrected by using the CAD road network data.

[0048] In another exemplary embodiment of this application, in order to accurately obtain the three-dimensional road network data of the underground tunnel, the target coordinates of any first road point are projected along the Z-axis onto the corresponding BIM model of the underground tunnel to obtain several BIM projection points, specifically including the following steps 401 to 402. Wherein: Step 401: Based on the number of any of the underground tunnels and the component attributes of the BIM model of the target mountain area, select the BIM component of any of the underground tunnels from the BIM components of the BIM model; wherein, by parsing the BIM model of the target mountain area, obtain each BIM component and component attributes of the BIM model.

[0049] For example, if the underground turnstile records that the current underground tunnel number is sd-001, then based on the component attributes of the BIM model (the attributes of the BIM components of the underground tunnel include the underground tunnel number), the BIM component with the number sd-001 is selected from the BIM model of the underground tunnel.

[0050] In this embodiment of the application, by filtering out the BIM components of any of the above-mentioned underground tunnels and eliminating unnecessary BIM components, the number of triangular meshes is reduced, and the projection speed of step 402 is improved.

[0051] Step 402: Project the target coordinates of any of the first road points along the negative Z-axis onto the BIM component of any of the underground tunnels to obtain several BIM projection points.

[0052] In another exemplary embodiment of this application, in order to accurately obtain the three-dimensional road network data of non-underground tunnel roads, the target coordinates of any second road point are projected along the Z-axis onto the GIS real-world model of the target mountain area to obtain several GIS projection points, specifically including the following steps 501 to 502. Wherein: Step 501: Based on the quadtree spatial index structure, according to the target coordinates of any second road point, retrieve the set of triangular faces within the spatial region where the second road point is located, as a subset of candidate triangular faces; wherein, the quadtree spatial index structure is obtained by parsing the above GIS real-world model to obtain the set of triangular face grids of the GIS real-world model, and based on the planar coordinates of the vertices of the triangular faces of the GIS real-world model ( x , y The spatial index tree of the triangular mesh quadtree is obtained by constructing the spatial index tree of the triangular mesh quadtree.

[0053] In this embodiment, a two-dimensional quadtree is a common data structure for improving spatial query efficiency. It divides the triangular mesh set of the GIS real-world model into four sub-regions, and further recursively subdivides them based on the number or spatial distribution characteristics of triangular meshes within each sub-region until a preset threshold condition is met. Each quadtree node stores the triangular mesh set within its coverage area, thus enabling rapid location of the triangular mesh set within the spatial region of any second road point during spatial queries, and allowing for accurate calculations within a local area, thereby improving overall processing efficiency.

[0054] Step 502: Project the target coordinates of any of the second road points along the negative Z-axis onto the candidate triangular face subset to obtain several GIS projection points.

[0055] In this embodiment of the application, projecting the target coordinates of any second road point along the negative Z-axis to the aforementioned candidate triangular face subset means that for any second road point... target coordinates Based on the maximum Z-value of all triangular faces in the GIS reality model, construct the corresponding 3D points. ( (Z), use a little more ( A ray is constructed along the negative Z-axis, and the intersection of the constructed ray with the candidate triangular face subset is detected to obtain several intersection points and a set of triangular faces. Each intersection point is a GIS projection point.

[0056] In another exemplary embodiment of this application, the above-described three-dimensional road network extraction method based on engineering models and vehicle positioning information further includes, before step 102: Step 601: Convert the acquired vehicle positioning data and coordinates of the target mountain area's GIS real-scene model to the right-hand rectangular two-dimensional coordinate system (XY coordinate system) where the BIM model of the target mountain area is located, and convert the converted two-dimensional coordinates of the vehicle positioning data of the target mountain area (XY coordinate system). x , y ) as the planar coordinates of the vehicle positioning data, and the two-dimensional coordinates after converting the GIS real-scene model of the target mountain area ( x , y () as the planar coordinates of the GIS reality model.

[0057] In another exemplary embodiment of this application, the above-described three-dimensional road network extraction method based on engineering models and vehicle positioning information further includes steps 602 to 603 before step 102 and after step 601, wherein: Step 602: Based on the gate information of various underground tunnel entrances in the target mountainous area, obtain the time range in which the target vehicle is located in each underground tunnel; wherein, the gate information also includes the time when the target vehicle enters and exits each underground tunnel.

[0058] In this embodiment of the application, the time range in which the target vehicle is located in each underground tunnel can be obtained by measuring the time the target vehicle enters and exits each underground tunnel.

[0059] Step 603: Based on the above time range, the vehicle positioning data of the target vehicle in the target mountain area is screened to obtain the vehicle positioning data of the target vehicle in various underground tunnels in the target mountain area and the vehicle positioning data of the target vehicle on non-underground tunnel roads in the target mountain area.

[0060] In this embodiment of the application, the vehicle positioning data of the target vehicle within the underground tunnel is filtered by combining the time range of the target vehicle being located in various underground tunnels with the timestamp of the vehicle positioning data, as well as the vehicle positioning data of the target vehicle on non-underground tunnel roads (outside the time range of the target vehicle being inside the underground tunnel). Furthermore, based on the road number of the non-underground tunnel road, the vehicle positioning data of each road segment of the non-underground tunnel road is obtained.

[0061] In another exemplary embodiment of this application, in order to ensure that the projected points are general road surfaces, the above-mentioned three-dimensional road network extraction method based on engineering models and vehicle positioning information further includes steps 701 to 703, wherein: Step 701: Calculate the normal vector of the triangle face containing the BIM projection point with the lowest Z coordinate.

[0062] Step 702: Determine the angle between the above-mentioned normal vector and the vertical direction. .

[0063] Step 703, detect the above included angle. If the angle is less than a preset threshold, then any first road point is determined to be a first abnormal road point and is removed.

[0064] In this embodiment of the application, since the slope of underground roads generally does not exceed 15° during construction, the angle threshold is set to 15°. If the included angle... If the angle is not less than the preset threshold, it may be projected onto the side of the underground tunnel or other parts, so this point needs to be removed.

[0065] In another exemplary embodiment of this application, in order to further ensure the accuracy of the extracted three-dimensional road network data, the above-mentioned three-dimensional road network extraction method based on engineering models and vehicle positioning information further includes steps 801 to 802, wherein: Step 801: Sort the three-dimensional coordinates of all road points according to the timestamp to obtain the three-dimensional data of the first target; wherein, the road points include the first road point and the second road point.

[0066] Step 802: Remove the second abnormal road points from the three-dimensional data of the first target to obtain the three-dimensional data of the third target, and fill the missing road point data in the three-dimensional data of the third target using linear interpolation; wherein, the second abnormal road points include road points with abnormal speed and road points with abnormal slope.

[0067] In another exemplary embodiment of this application, step 802 above, removing the second abnormal road point from the three-dimensional data of the first target, specifically includes steps 8021 to 8022, wherein: Step 8021: According to the timestamp of each road point in the first target 3D data, perform speed processing operation on each road point in the first target 3D data as the first target point in sequence, until the second to last road point in the first target 3D data is used as the first target point and the speed processing operation is completed, to obtain the second target 3D data.

[0068] The speed processing operation includes the following steps (1) to (3): Step (1): Calculate the Euclidean distance between the first target point and the adjacent road point.

[0069] In this embodiment of the application, the adjacent road points of the first target point refer to the road points in the three-dimensional data of the first target that are adjacent to the first target point.

[0070] First target point With adjacent road points Euclidean distance Calculated using the following formula: .

[0071] Step (2) is to obtain the vehicle speed by dividing the Euclidean distance between the first target point and the adjacent road point by the sampling time interval between the first target point and the adjacent road point.

[0072] In this embodiment, the sampling time interval between the first target point and the adjacent road point is... It is obtained through the timestamps of adjacent road points and the first target point.

[0073] Vehicle speed Calculated using the following formula: .

[0074] Step (3): Detect whether the vehicle's moving speed exceeds the set speed threshold. If so, then the adjacent road points of the aforementioned first target point are considered to be... Points with abnormal speeds are removed.

[0075] Step 8022: According to the timestamp of each road point in the second target 3D data, perform slope processing operation on each road point in the second target 3D data as the second target point in turn, until the second to last road point in the second target 3D data is used as the second target point and the slope processing operation is completed, to obtain the third target 3D data.

[0076] The slope treatment operation includes the following steps a to c, wherein: Step a: Calculate the horizontal distance between the second target point and the adjacent road point.

[0077] In this embodiment of the application, the second target point With adjacent road points The horizontal distance is calculated using the following formula: .

[0078] Step b, based on the above horizontal distance Calculate the road slope between the second target point and the adjacent road points by using the Z coordinates of the second target point and the Z coordinates of the adjacent road points.

[0079] In this embodiment of the application, the road slope between the second target point and the adjacent road point is... Calculated using the following formula: .

[0080] Step c, Detect the slope of the above-mentioned road Is it greater than the set slope threshold? If so, then the adjacent road points of the second target point are considered. Points with abnormal slopes are removed.

[0081] In another exemplary embodiment of this application, step 802 above, filling in the missing road point data in the three-dimensional data of the third target using linear interpolation, specifically includes: like For the second abnormal road point, linear interpolation was used in the three-dimensional data of the third target. Road point data that fills gaps between adjacent valid 3D points.

[0082] In this embodiment of the application, adjacent valid three-dimensional points refer to the three-dimensional data of the third target, and... Adjacent road points.

[0083] In this embodiment of the application, invalid data in the acquired vehicle positioning data of the target mountain area is removed, that is, vehicle positioning data in the acquired vehicle positioning data of the target mountain area with signal strength lower than a preset strength threshold and / or accuracy factor not less than a preset accuracy factor threshold is removed.

[0084] The acquired vehicle positioning data includes parameters such as satellites (number of satellites), signal intensity (signal strength), and precision (precision factor) to determine the accuracy of the data. When the number of satellites is greater than 6, the signal strength is high, and the precision factor is less than 2.5, the data can be considered valid. Therefore, vehicle positioning data with signal strength lower than a preset threshold in the acquired target mountain area are removed, that is, vehicle positioning data with no more than 6 satellites and / or a precision factor of not less than 2.5 are removed.

[0085] Existing methods for verifying the continuity and integrity of mountain road data do not incorporate multi-source positioning data and engineering models (BIM models and GIS reality models), leading to anomalies in the extracted 3D road network data, reducing its quality, and affecting its reliability. This application's embodiment, however, combines BIM models and GIS reality models to extract 3D road network data, ensuring continuity between mountain roads and underground tunnel roads. It also employs speed and slope thresholds to filter out unreasonable road points, further guaranteeing the quality and reliability of the extracted 3D road network data.

[0086] In another exemplary embodiment of this application, the above-described three-dimensional road network extraction method based on engineering models and vehicle positioning information further includes steps 901 to 909 after step 802, wherein: Step 901: Discretize the three-dimensional data of the third target at the meter level to obtain discretized road points, and construct a w×w two-dimensional grid road covering all valid road points in the three-dimensional data of the third target, where w is the width.

[0087] In this embodiment of the application, the three-dimensional data of the third target is discretized at the meter level. That is, the road points in the three-dimensional data of the third target are resampled and discretized according to a preset distance interval. The distance interval is not specifically limited and can be set according to actual needs. The preferred range for setting the distance interval is 0.5m-3.0m.

[0088] Step 902: Map all discrete road points to the corresponding grids of the two-dimensional grid roads, and count the number of roads covered by the discrete road points in each grid of the two-dimensional grid roads to obtain the number of roads covered by each grid.

[0089] Step 903: Check if the number of roads covered by each grid is less than 2.

[0090] Step 904: If any grid cell covers a road with a number of roads less than 2 and a number of roads equal to 1, then the road is determined to be a regular road segment.

[0091] Step 905: If the number of roads covered by any grid is not less than 2, then detect whether there are at least two roads in the roads covered by the grid that have an intersection area and the elevation difference is less than the set first elevation difference threshold. If so, then it is considered that there is an intersection segment in the grid and the grid is determined to be an intersection area.

[0092] In this embodiment of the application, the elevation difference between any two roads can be selected as the average value of the Z coordinates of the effective road points in the intersection area of ​​the two roads.

[0093] Step 906: Extract the valid road points of the intersection segments in each intersection area, and calculate the geometric center point of the extracted valid road points as the intersection breakpoint.

[0094] Step 907: Divide the intersection area into segments at each intersection breakpoint and remove redundant and overlapping road points at each intersection breakpoint to obtain independent road segments that do not intersect and are connected only at each intersection breakpoint.

[0095] Step 908: Convert the planar coordinates of the valid road points of ordinary road segments and independent road segments to the WGS-84 coordinate system and / or CGCS2000 coordinate system, and encapsulate them into GeoJSON, SHP, DXF or 3D model format for output.

[0096] In this embodiment, it is not necessary to perform coordinate transformation on the elevation of valid road points for ordinary road sections and independent road sections.

[0097] In another exemplary embodiment of this application, the above-described three-dimensional road network extraction method based on engineering models and vehicle positioning information further includes, before step 601: Invalid data in the acquired vehicle positioning data of the target mountainous area is removed, and the retained vehicle positioning data is arranged in order of timestamp.

[0098] In another exemplary embodiment of this application, the vehicle positioning data and GIS real-scene model data of the target mountain area in the WGS-84 coordinate system or CGCS2000 coordinate system are uniformly converted to the right-hand rectangular coordinate system where the BIM model of the target mountain area is located by the open-source library Proj.4.

[0099] In another exemplary embodiment of this application, a vehicle-mounted GPS device is used to collect positioning data of non-underground tunnel roads in the target mountainous area to obtain vehicle-mounted positioning data of underground tunnel roads in the target mountainous area. A UWB ultra-wideband positioning module is used to collect positioning data of underground tunnel roads in the target mountainous area. The UWB ultra-wideband positioning module includes base stations installed in the underground tunnels and UWB tags installed on the target vehicle.

[0100] In another exemplary embodiment of this application, the above-described 3D road network extraction method based on engineering models and vehicle positioning information is applied to the 3D road network extraction of a mountainous area surrounding a dam. Vehicle GPS devices and UWB tags are used to collect vehicle positioning data of the mountainous area surrounding the dam at a frequency of 1 time / second, along with CAD road network data and gate information for various underground tunnel entrances. Using Proj.4, the WGS-84 format vehicle positioning data and GIS real-scene model data are uniformly converted to the right-hand rectangular coordinate system of the BIM model, retaining the X and Y two-dimensional coordinates to obtain the point set P(X,Y). Vehicle positioning data L1 (1268 positioning points) for underground tunnel roads and vehicle positioning data L2 (8942 positioning points) for non-underground tunnel roads are filtered based on gate information. Three BIM components for underground tunnels in the BIM model are selected. Combining gate information and component attributes of the BIM model, corresponding CAD road network data are selected, and the positioning points of L1 are traversed. Vertical projection correction is performed on each positioning point of L1 according to a distance threshold D1=1m. The corrected L1 positioning points are projected along the negative Z-axis to the corresponding BIM components. The road points with the lowest Z-coordinates are selected, and road points with an angle θ ≥ 15° between the normal vector and the vertical direction are removed, retaining 1226 valid road points. CAD road network data for non-underground tunnel roads is then filtered. Each positioning point of L2 is traversed, and points are corrected according to the D1=1m threshold. The corresponding GIS triangular facets for each positioning point of L2 are retrieved using a quadtree index. The road points with the lowest Z-coordinates are selected by projecting along the negative Y-axis. Combined with the valid road points of L1, the 3D road network data (10168 points) is obtained. =20km / h threshold, 108 abnormal distance and speed points were removed from consecutive points in L1 and L2. A 15% slope threshold was used to remove 76 outlier points, and linear interpolation was used to complete 68 missing points, resulting in 9992 valid 3D points. These valid points were discretized at the meter level to construct a 4m×4m 2D grid. The discrete points were mapped to the grid, identifying 6 intersection areas (e.g., 3 crossroads, 3 T-junctions), 572 ordinary road segments, and 3 intersection areas at different levels. Points in the intersection areas were extracted, and the geometric center points were calculated as breakpoints, allowing for the division into 21 independent road segments. 132 overlapping points were removed. The final valid road points were converted to the CGCS2000 coordinate system using Proj.4 and output in GeoJSON, SHP, and DXF formats. The entire process took approximately 2.5 hours, with an accuracy of ±0.8m, meeting engineering requirements.

[0101] This application proposes a method for extracting a 3D road network based on engineering models and vehicle positioning information. This method combines BIM models and vehicle positioning data to achieve rapid modeling of underground tunnel roads; it also combines GIS real-world models and vehicle positioning data to achieve rapid modeling of non-underground tunnel roads around dams; the method uses CAD drawings to correct road network data and establishes rules for road slope and speed verification to clean the data; it combines BIM attributes and quadtree structures to effectively improve modeling efficiency; and it uses grid cells to identify intersections and remove duplicate roads. This method is adaptable to the complex terrain of mountainous areas around dams, the enclosed environment of underground caverns, and the special needs of temporary roads, ensuring the efficiency, integrity, and reliability of the 3D road network construction.

[0102] Based on the same inventive concept, this application also provides a 3D road network extraction system based on engineering models and vehicle positioning information for implementing the aforementioned 3D road network extraction method based on engineering models and vehicle positioning information. The solution provided by this system is similar to the implementation scheme described in the above method. Therefore, the specific limitations of one or more embodiments of the 3D road network extraction system based on engineering models and vehicle positioning information provided below can be found in the limitations of the 3D road network extraction method based on engineering models and vehicle positioning information described above, and will not be repeated here.

[0103] In one exemplary embodiment, such as Figure 2 As shown, a three-dimensional road network extraction system 100 based on engineering models and vehicle positioning information is provided, including: The vehicle-mounted GPS device 1001 is used to collect vehicle positioning data on non-underground tunnel roads in the target mountainous area; The UWB ultra-wideband positioning module 1002 is used to collect vehicle positioning data of underground tunnels and roads in the target mountainous area. Processing module 1003 is used for: The system acquires vehicle positioning data collected by the vehicle-mounted GPS device 1001 and the UWB ultra-wideband positioning module 1002; based on the planar coordinates of the vehicle positioning data of any first road point, it performs BIM model projection to obtain several BIM projection points, and based on the planar coordinates of the vehicle positioning data of any second road point, it performs GIS real-scene model projection to obtain several GIS projection points; wherein, the first road point is any positioning point in the vehicle positioning data of any underground tunnel road, and the vehicle positioning data of any underground tunnel road is collected by the positioning module in the underground tunnel; the second road point is any positioning point in the vehicle positioning data of any section of any non-underground tunnel road; the system selects the three-dimensional coordinates of the BIM projection point with the lowest Z coordinate as the three-dimensional coordinates of the first road point, and selects the three-dimensional coordinates of the GIS projection point with the lowest Z coordinate as the three-dimensional coordinates of the second road point, thereby obtaining the three-dimensional road network data of the target mountain area.

[0104] In this embodiment, by acquiring vehicle positioning data of the target mountain area, and based on the planar coordinates of the vehicle positioning data (collected by the positioning module inside the underground tunnel) of any first road point (any positioning point in the vehicle positioning data of any underground tunnel road), BIM model projection is performed to obtain several BIM projection points. Then, based on the planar coordinates of the vehicle positioning data of any second road point, GIS real-scene model projection is performed to obtain several GIS projection points. The three-dimensional coordinates of the BIM projection point with the lowest Z-coordinate are selected as the three-dimensional coordinates of the first road point, and the three-dimensional coordinates of the GIS projection point with the lowest Z-coordinate are selected as the three-dimensional coordinates of the second road point. This yields the three-dimensional road network data of the target mountain area, enabling comprehensive extraction of three-dimensional road network data for both underground tunnel roads and non-underground tunnel roads in the target mountain area. This solves the problem of GPS signal loss in environments such as underground tunnels, where existing GPS positioning combined with elevation fitting methods cannot obtain three-dimensional road data for underground caverns and other environments, resulting in incomplete three-dimensional road data.

[0105] Meanwhile, by combining vehicle-mounted positioning data with BIM and GIS reality models, the obtained elevation coordinates of underground and non-underground tunnel roads are more accurate, thus solving the problem of insufficient elevation positioning accuracy caused by GPS positioning combined with elevation fitting. Furthermore, compared to existing methods that extract 3D road data solely based on BIM models, the use of GIS reality models avoids the low accuracy problem caused by poor real-time performance in BIM model-based 3D road data extraction, resulting in better real-time performance and thus more accurate data. In addition, the 3D road network data obtained through vehicle-mounted positioning data has better real-time performance and is more comprehensive and accurate than traditional CAD road network data, solving the problems of poor real-time performance and low accuracy in 3D road data extracted from CAD road network data. Therefore, the embodiments of this application can accurately obtain 3D road data of the mountainous area surrounding the dam.

[0106] In summary, the three-dimensional road data extraction method of this application embodiment can accurately and comprehensively obtain three-dimensional road data of the mountainous area surrounding the dam.

[0107] As an optional implementation, the processing module 1003 is further configured to: While acquiring vehicle positioning data of the target mountain area, the system also acquires CAD road network data of the target mountain area and gate information of various underground tunnel entrances. The CAD road network data includes road coordinates and road attributes, and the road attributes include underground tunnels and non-underground tunnels. The gate information includes at least the number of the underground tunnel.

[0108] As an optional implementation, the processing module 1003 is further configured to: Before projecting a BIM model based on the planar coordinates of the vehicle positioning data of any first road point to obtain several BIM projection points, and before projecting a GIS real-scene model based on the planar coordinates of the vehicle positioning data of any second road point to obtain several GIS projection points, the system checks whether any road point has corresponding CAD road network data (including at least two-dimensional coordinate data) based on the road attributes of the target mountain area and the gate information of each underground tunnel entrance; wherein, the road point includes the first road point and the second road point.

[0109] Accordingly, regarding the BIM model projection based on the planar coordinates of the vehicle positioning data of any first road point to obtain several BIM projection points, and the GIS reality model projection based on the planar coordinates of the vehicle positioning data of any second road point to obtain several GIS projection points, the above processing module 1003 is specifically used for: If any of the above road points has corresponding CAD road network data, perform vertical projection correction and engineering model projection operations in sequence; if any of the above road points does not have corresponding CAD road network data, perform engineering model projection operations.

[0110] The vertical projection correction operation includes: Based on the CAD road network data of any of the above road points, the planar coordinates of the vehicle positioning data of any of the above road points are ( x , y Perform planar distance threshold correction to obtain the corrected coordinates of any road point.

[0111] Engineering model projection operations include: Projecting the target coordinates of any first road point along the Z-axis onto the BIM model of the corresponding underground tunnel yields several BIM projection points; and / or projecting the target coordinates of any second road point along the Z-axis onto the GIS real-world model of the target mountain area yields several GIS projection points; wherein the target coordinates are calibrated coordinates or planar coordinates of vehicle positioning data.

[0112] Here, projecting the target coordinates of any first road point along the Z-axis onto the BIM model of any of the aforementioned underground tunnels means that for any first road point... target coordinates Based on the maximum Z-value of all triangular faces in the BIM model, construct the corresponding 3D points. ( (Z), use a little more ( ,Z) Construct a ray along the negative Z-axis, and perform intersection detection between the constructed ray and the triangular mesh model of the corresponding underground tunnel BIM component to obtain several intersection points and the corresponding triangular face set. Each intersection point is a BIM projection point.

[0113] As an optional implementation, for the CAD road network data based on any of the above road points, the planar coordinates of the vehicle positioning data of any of the road points are ( x , y The above processing module 1003 is specifically used to perform planar distance threshold correction to obtain the aspect of the corrected coordinates of any road point. Calculate the planar coordinates of the vehicle positioning data for any of the above road points. x , y Find the shortest distance to the corresponding CAD road network data segment and check if the shortest distance is less than a preset distance threshold. If so, project any road point onto the CAD road network data segment along a direction perpendicular to the CAD road network data segment to obtain the corrected coordinates of any road point.

[0114] As an optional implementation, regarding the projection of the target coordinates of any first road point along the Z-axis onto the BIM model of the corresponding underground tunnel to obtain several BIM projection points, the aforementioned processing module 1003 is specifically used for: Based on the number of any of the underground tunnels and the component attributes of the BIM model of the target mountain area, the BIM components of any of the underground tunnels are selected from the BIM components of the BIM model; wherein, by parsing the BIM model of the target mountain area, the BIM components and component attributes of the BIM model are obtained; the target coordinates of any of the first road points are projected along the negative Z-axis to the BIM components of any of the underground tunnels to obtain several BIM projection points.

[0115] In this embodiment of the application, projecting the target coordinates of any second road point along the negative Z-axis to the aforementioned candidate triangular face subset means that for any second road point... target coordinates Based on the maximum Z-value of all triangular faces in the GIS reality model, construct the corresponding 3D points. ( (Z), use a little more ( A ray is constructed along the negative Z-axis, and the intersection of the constructed ray with the candidate triangular face subset is detected to obtain several intersection points and a set of triangular faces. Each intersection point is a GIS projection point.

[0116] As an optional implementation, regarding the projection of the target coordinates of any second road point along the Z-axis onto the GIS real-world model of the target mountain area to obtain several GIS projection points, the aforementioned processing module 1003 is specifically used for: Based on the quadtree spatial index structure, according to the target coordinates of any second road point, the set of triangular faces within the spatial region where the second road point is located is retrieved as a subset of candidate triangular faces; wherein, the quadtree spatial index structure is obtained by parsing the above GIS reality model to obtain the set of triangular face meshes of the GIS reality model, and based on the planar coordinates of the vertices of the triangular faces of the GIS reality model ( x , y Construct the quadtree spatial index tree of the triangular mesh to obtain the quadtree spatial index tree of the triangular mesh; project the target coordinates of any of the second road points along the negative Z-axis to the above candidate triangular face subset to obtain several GIS projection points.

[0117] As an optional implementation, the processing module 1003 is further configured to: Before projecting the BIM model using the planar coordinates of the vehicle positioning data from any first road point to obtain several BIM projection points, and before projecting the GIS reality model using the planar coordinates of the vehicle positioning data from any second road point to obtain several GIS projection points, the coordinates of the acquired vehicle positioning data and the GIS reality model of the target mountain area are uniformly converted to the right-hand rectangular two-dimensional coordinate system (XY coordinate system) where the BIM model of the target mountain area is located. The converted two-dimensional coordinates of the vehicle positioning data of the target mountain area are then... x , y ) as the planar coordinates of the vehicle positioning data, and the two-dimensional coordinates after converting the GIS real-scene model of the target mountain area ( x , y () as the planar coordinates of the GIS reality model.

[0118] As an optional implementation, the processing module 1003 is further configured to: Before projecting the BIM model using the planar coordinates of the vehicle positioning data from any first road point to obtain several BIM projection points, and before projecting the GIS reality model using the planar coordinates of the vehicle positioning data from any second road point to obtain several GIS projection points, the coordinates of the acquired vehicle positioning data and the GIS reality model of the target mountain area are uniformly converted to the right-hand rectangular two-dimensional coordinate system of the BIM model of the target mountain area. Based on the gate information of each underground tunnel entrance in the target mountain area, the time range of the target vehicle being located in each underground tunnel is obtained. The gate information also includes the time of the target vehicle entering and exiting each underground tunnel. Based on the above time range, the vehicle positioning data of the target vehicle in the target mountain area is filtered to obtain the vehicle positioning data of the target vehicle in each underground tunnel in the target mountain area and the vehicle positioning data of the target vehicle on non-underground tunnel roads in the target mountain area.

[0119] As an optional implementation, the processing module 1003 is further configured to: Calculate the normal vector of the triangular face containing the BIM projection point with the lowest Z-coordinate; determine the angle between the normal vector and the vertical direction. ; Detect the above included angle If the angle is less than a preset threshold, then any first road point is determined to be a first abnormal road point and is removed.

[0120] As an optional implementation, the processing module 1003 is further configured to: Sort the 3D coordinates of all road points according to timestamps to obtain the first target 3D data; the road points include the first road point and the second road point; remove the second abnormal road points in the first target 3D data to obtain the third target 3D data, and fill the missing road point data in the third target 3D data by linear interpolation; the second abnormal road points include road points with abnormal speed and road points with abnormal slope.

[0121] As an optional implementation, regarding the removal of second abnormal road points from the first target's three-dimensional data, the processing module 1003 is specifically used for: Based on the timestamp of each road point in the first target 3D data, perform speed processing operations on each road point in the first target 3D data as the first target point, until the second-to-last road point in the first target 3D data is used as the first target point and the speed processing operation is completed, to obtain the second target 3D data; based on the timestamp of each road point in the second target 3D data, perform slope processing operations on each road point in the second target 3D data as the second target point, until the second-to-last road point in the second target 3D data is used as the second target point and the slope processing operation is completed, to obtain the third target 3D data.

[0122] For details regarding speed processing and slope processing operations, please refer to the descriptions in the above method embodiments; they will not be repeated here.

[0123] As an optional implementation, the processing module 1003 is further configured to: After removing the second abnormal road points from the first target's 3D data to obtain the third target's 3D data, and filling in the missing road point data in the third target's 3D data using linear interpolation, the third target's 3D data is discretized at the meter level to obtain discretized road points. A w×w two-dimensional grid road network is then constructed, covering all valid road points in the third target's 3D data, where w is the width. All discretized road points are mapped to the corresponding grids of the above two-dimensional grid road network, and the number of roads covered by the discretized road points in each grid of the two-dimensional grid road network is counted to obtain the number of roads covered by each grid. It is then checked whether the number of roads covered by each grid is less than 2. If the number of roads covered by any grid is less than 2 and is 1, then the road is determined to be a normal road segment. If the number of roads covered by any grid is not less than 2, then the road segment is determined to be normal. For step 2, it checks whether there are at least two roads intersecting within the roads covered by any given grid, and whether the elevation difference is less than a set first elevation difference threshold. If so, it considers that there is an intersection segment within any given grid and determines that given grid as an intersection area. It extracts the valid road points of the intersection segments in each intersection area and calculates the geometric center point of the extracted valid road points as intersection breakpoints. It splits the intersection segments of the corresponding intersection area from each intersection breakpoint and removes redundant and overlapping road points at each intersection breakpoint to obtain independent road segments that do not intersect and are connected only at each intersection breakpoint. It transforms the planar coordinates of the valid road points of ordinary road segments and independent road segments to the WGS-84 coordinate system and / or CGCS2000 coordinate system, and encapsulates them into GeoJSON, SHP, DXF, or 3D model format for output.

[0124] As an optional implementation, the processing module 1003 is further configured to: Before uniformly converting the acquired vehicle positioning data and the coordinates of the GIS real-scene model of the target mountain area to the right-hand rectangular two-dimensional coordinate system (XY coordinate system) where the BIM model of the target mountain area is located, invalid data in the acquired vehicle positioning data of the target mountain area is removed, and the retained vehicle positioning data is arranged in the order of timestamps.

[0125] In this embodiment of the application, invalid data in the acquired vehicle positioning data of the target mountain area is removed, that is, vehicle positioning data in the acquired vehicle positioning data of the target mountain area with signal strength lower than a preset strength threshold and / or accuracy factor not less than a preset accuracy factor threshold is removed.

[0126] The acquired vehicle positioning data includes parameters such as satellites (number of satellites), signal intensity (signal strength), and precision (precision factor) to determine the accuracy of the data. When the number of satellites is greater than 6, the signal strength is high, and the precision factor is less than 2.5, the data can be considered valid. Therefore, vehicle positioning data with signal strength lower than a preset threshold in the acquired target mountain area are removed, that is, vehicle positioning data with no more than 6 satellites and / or a precision factor of not less than 2.5 are removed.

[0127] As an optional implementation, regarding the unified conversion of the acquired vehicle positioning data and coordinates of the GIS real-scene model of the target mountain area to the right-hand rectangular two-dimensional coordinate system of the BIM model of the target mountain area, the above-mentioned processing module 1003 is specifically used for: The open-source library Proj.4 is used to uniformly convert vehicle positioning data and GIS reality model data of the target mountain area in the WGS-84 or CGCS2000 coordinate system to the right-hand rectangular coordinate system of the BIM model of the target mountain area.

[0128] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 3 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores data for a 3D road network extraction method based on engineering models and vehicle positioning information. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a 3D road network extraction method based on engineering models and vehicle positioning information.

[0129] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0130] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0131] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0132] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0133] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0134] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0135] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0136] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for extracting a three-dimensional road network based on an engineering model and vehicle positioning information, characterized in that, The method for extracting a 3D road network based on engineering models and vehicle positioning information includes: Obtain vehicle location data for the target mountainous area; Based on the planar coordinates of the vehicle positioning data of any first road point, a BIM model is projected to obtain several BIM projection points. Based on the planar coordinates of the vehicle positioning data of any second road point, a GIS real-scene model is projected to obtain several GIS projection points. The first road point is any positioning point in the vehicle positioning data of any underground tunnel road, and the vehicle positioning data of any underground tunnel road is collected by a positioning module in the underground tunnel. The second road point is any positioning point in the vehicle positioning data of any section of any non-underground tunnel road. The three-dimensional coordinates of the BIM projection point with the lowest Z-coordinate are selected as the three-dimensional coordinates of any first road point, and the three-dimensional coordinates of the GIS projection point with the lowest Z-coordinate are selected as the three-dimensional coordinates of any second road point, so as to obtain the three-dimensional road network data of the target mountain area. The process of projecting a BIM model based on the planar coordinates of vehicle positioning data from any first road point to obtain several BIM projection points, and projecting a GIS reality model based on the planar coordinates of vehicle positioning data from any second road point to obtain several GIS projection points, specifically includes: If any of the road points has corresponding CAD road network data, the vertical projection correction operation and the engineering model projection operation are performed sequentially; wherein, the road points include the first road point and the second road point; If no corresponding CAD road network data exists for any of the road points, perform an engineering model projection operation; The vertical projection correction operation includes: Based on the CAD road network data of any road point, the planar coordinates of the vehicle positioning data of any road point are determined. x , y Perform planar distance threshold correction to obtain the corrected coordinates of any road point; The engineering model projection operation includes: Projecting the target coordinates of any first road point along the Z-axis onto the BIM model of the corresponding underground tunnel yields several BIM projection points; and / or projecting the target coordinates of any second road point along the Z-axis onto the GIS real-world model of the target mountain area yields several GIS projection points; wherein the target coordinates are calibrated coordinates or planar coordinates of vehicle positioning data.

2. The method for extracting a three-dimensional road network based on an engineering model and vehicle positioning information according to claim 1, characterized in that, While acquiring vehicle positioning data of the target mountain area, CAD road network data of the target mountain area and gate information of various underground tunnel entrances are also acquired; wherein, the CAD road network data includes road coordinates and road attributes, the road attributes include underground tunnels and non-underground tunnels, and the gate information includes at least the number of the underground tunnel. The method for extracting a 3D road network based on an engineering model and vehicle positioning information further includes, before projecting the BIM model based on the planar coordinates of the vehicle positioning data of any first road point: Based on the road attributes of the target mountainous area and the gate information of each underground tunnel entrance, detect whether there is corresponding CAD road network data for any road point.

3. The method for extracting a three-dimensional road network based on an engineering model and vehicle positioning information according to claim 2, characterized in that, The planar coordinates of the vehicle positioning data of any road point based on the CAD road network data of any road point ( x , y Perform planar distance threshold correction to obtain the corrected coordinates of any road point, specifically including: Calculate the planar coordinates of the vehicle positioning data for any of the road points. x , y Find the shortest distance to the corresponding CAD road network data segment, and detect whether the shortest distance is less than a preset distance threshold. If so, project any road point onto the CAD road network data segment along a direction perpendicular to the CAD road network data segment to obtain the corrected coordinates of any road point. Projecting the target coordinates of any first road point along the Z-axis onto the BIM model of any underground tunnel yields several BIM projection points, specifically including: Based on the number of any underground tunnel and the component attributes of the BIM model of the target mountain area, the BIM component of any underground tunnel is selected from the BIM components of the BIM model; wherein, by parsing the BIM model of the target mountain area, each BIM component and component attributes of the BIM model are obtained. Project the target coordinates of any first road point along the negative Z-axis onto the BIM component of any underground tunnel to obtain several BIM projection points. Projecting the target coordinates of any second road point along the Z-axis onto the GIS real-world model of the target mountain area yields several GIS projection points, specifically including: Based on the quadtree spatial index structure, according to the target coordinates of any second road point, the set of triangular faces within the spatial region where the second road point is located is retrieved as a subset of candidate triangular faces; wherein, the quadtree spatial index structure is used to parse the GIS real-world model to obtain a set of triangular face meshes of the GIS real-world model, and based on the planar coordinates of the vertices of the triangular faces of the GIS real-world model ( x , y The triangular mesh quadtree spatial index tree obtained by constructing the triangular mesh quadtree spatial index tree; Project the target coordinates of any second road point along the negative Z-axis onto the candidate triangular face subset to obtain several GIS projection points.

4. The method for extracting a three-dimensional road network based on an engineering model and vehicle positioning information according to claim 2, characterized in that, Before detecting whether corresponding CAD road network data exists for any road point based on the road attributes of the target mountainous area and the gate information of various underground tunnel entrances, the method further includes: The acquired vehicle positioning data and coordinates of the GIS real-scene model of the target mountain area are uniformly converted to the right-hand rectangular two-dimensional coordinate system of the BIM model of the target mountain area, and the converted two-dimensional coordinates of the vehicle positioning data of the target mountain area are then converted. x , y The planar coordinates of the vehicle positioning data are used as the two-dimensional coordinates after the GIS real-scene model of the target mountain area is converted. x , y () is used as the planar coordinate of the GIS real-scene model.

5. The method for extracting a three-dimensional road network based on an engineering model and vehicle positioning information according to claim 1, characterized in that, Before detecting whether corresponding CAD road network data exists for any road point based on the road attributes of the target mountainous area and the gate information of various underground tunnel entrances, the method further includes: Based on the gate information of various underground tunnel entrances in the target mountainous area, the time range of the target vehicle being located in each underground tunnel is obtained; wherein, the gate information also includes the time of the target vehicle entering and exiting each underground tunnel; Based on the time range, the vehicle positioning data of the target vehicle in the target mountain area is screened to obtain the vehicle positioning data of the target vehicle in various underground tunnels in the target mountain area and the vehicle positioning data of the target vehicle on non-underground tunnel roads in the target mountain area.

6. The method for extracting a three-dimensional road network based on an engineering model and vehicle positioning information according to claim 1, characterized in that, Also includes: Calculate the normal vector of the triangle face containing the BIM projection point with the lowest Z coordinate; Determine the angle between the normal vector and the vertical direction. ; Detect the included angle If the angle is less than a preset threshold, and if it is not less than a preset threshold, then any of the first road points is determined to be the first abnormal road point and is removed.

7. The method for extracting a three-dimensional road network based on an engineering model and vehicle positioning information according to claim 6, characterized in that, Also includes: Sort the three-dimensional coordinates of all road points according to their timestamps to obtain the three-dimensional data of the first target. The second abnormal road points in the first target's three-dimensional data are removed to obtain the third target's three-dimensional data, and the missing road point data in the third target's three-dimensional data are filled in by linear interpolation; wherein, the second abnormal road points include road points with abnormal speed and road points with abnormal slope.

8. The method for extracting a three-dimensional road network based on an engineering model and vehicle positioning information according to claim 7, characterized in that, The removal of the second abnormal road point from the three-dimensional data of the first target specifically includes: Based on the timestamp of each road point in the first target 3D data, each road point in the first target 3D data is sequentially treated as a first target point and speed processing is performed until the second to last road point in the first target 3D data is treated as a first target point and the speed processing operation is completed, thus obtaining the second target 3D data. According to the timestamp of each road point in the second target 3D data, the slope processing operation is performed on each road point in the second target 3D data as the second target point in turn, until the slope processing operation is performed on the second to last road point in the second target 3D data as the second target point, and the third target 3D data is obtained. The speed processing operation includes: Calculate the Euclidean distance between the first target point and the adjacent road points; The vehicle speed is obtained by dividing the Euclidean distance by the sampling time interval between the first target point and the adjacent road point; If the vehicle's moving speed is greater than a set speed threshold, then the adjacent road points of the first target point are considered to be speed anomaly points and are removed. The slope processing operation includes: Calculate the horizontal distance between the second target point and the adjacent road point; Based on the horizontal distance, the Z-coordinate of the second target point and the Z-coordinate of the adjacent road point of the second target point, calculate the road slope between the second target point and the adjacent road point; If the road slope is greater than the set slope threshold, the adjacent road points of the second target point are considered to be slope abnormal points and are removed.

9. The method for extracting a three-dimensional road network based on an engineering model and vehicle positioning information according to claim 8, characterized in that, After removing the second abnormal road points from the first target's 3D data and filling in the removed second abnormal road points using linear interpolation, the process further includes: The three-dimensional data of the third target is discretized at the meter level to obtain discretized road points, and a w×w two-dimensional grid road covering all valid road points in the three-dimensional data of the third target is constructed, where w is the width; Map all discrete road points to the corresponding grids of the two-dimensional grid roads, and count the number of roads covered by the discrete road points in each grid of the two-dimensional grid roads to obtain the number of roads covered by each grid. Check if the number of roads covered by each grid is less than 2; If any grid cell covers a road segment with a number of roads less than 2 and a number of roads equal to 1, then the road segment is determined to be a regular road segment. If the number of roads covered by any grid is not less than 2, then it is detected whether there are at least two roads in the road covered by the grid that have an intersection area and the height difference is less than the set first height difference threshold. If so, it is considered that there is an intersection segment in the grid and the grid is determined to be an intersection area. Extract the valid road points of the intersection segments in each intersection area, and calculate the geometric center point of the extracted valid road points as the intersection breakpoint; The intersection area is split from each intersection breakpoint, and redundant and overlapping road points at each intersection breakpoint are removed to obtain independent road segments that do not intersect and are connected only at the intersection breakpoints. The planar coordinates of the valid road points of the ordinary road segments and independent road segments are transformed to the WGS-84 coordinate system and / or CGCS2000 coordinate system, and then encapsulated as GeoJSON, SHP, DXF or 3D model format for output.

10. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the three-dimensional road network extraction method based on an engineering model and vehicle positioning information as described in any one of claims 1-9.

Images