A road structure identification method, device and equipment and readable storage medium

Abstract

The application discloses a road structure identification method, device and equipment and a readable storage medium. The method comprises the following steps: acquiring a two-dimensional map and elevation data corresponding to a road of a preset type in a target area; determining a two-dimensional model of each road of the road of the preset type in the target area according to the two-dimensional map; determining a three-dimensional model of each road according to the two-dimensional model of each road and the corresponding elevation data; determining a three-dimensional road feature of each road according to the three-dimensional model of each road; and obtaining a road structure in the target area according to the three-dimensional road feature of each road and the relationship between the roads. The above technical solution disclosed by the application can improve the accuracy of road structure identification by establishing a three-dimensional model of a road, determining a three-dimensional road feature according to the three-dimensional model and obtaining a road structure according to the three-dimensional road feature, thereby better providing a road basis for automatic driving research of a vehicle.

Description

Technical Field

[0001] This application relates to the field of road recognition technology, and more specifically, to a road structure recognition method, apparatus, device, and readable storage medium. Background Technology

[0002] Fully autonomous driving in the long term requires sensing passenger intentions, vehicle conditions, road traffic conditions, and other factors, and analyzing, predicting, and judging these conditions to provide passengers with a safe and comfortable experience and to achieve efficient operation of the transportation system. Vehicle-to-everything (V2X) communication with edge intelligence can provide ultra-low latency information sharing and enhanced data analysis and processing capabilities between vehicles, making fully autonomous driving possible.

[0003] Currently, existing edge computing for vehicle-to-everything (V2X) systems primarily uses vehicle motion models that move along straight roads or remain stationary on a plane during execution. These models are two-dimensional maps, meaning that road structure identification is based on two-dimensional maps. However, two-dimensional maps lose spatial information and cannot fully reflect the objective world. Therefore, the accuracy of road structure identification based on this model is not high, and it cannot better provide road constraints for network communication and resource allocation in V2X systems.

[0004] In conclusion, improving the accuracy of road structure identification is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] In view of this, the purpose of this application is to provide a road structure identification method, apparatus, device and readable storage medium to improve the accuracy of road structure identification.

[0006] To achieve the above objectives, this application provides the following technical solution:

[0007] A road structure identification method, comprising:

[0008] Obtain the 2D map and elevation data corresponding to roads of preset types within the target area;

[0009] Based on the two-dimensional map, determine the two-dimensional model of each road of the preset type within the target area;

[0010] Based on the two-dimensional models of each road and the corresponding elevation data, the three-dimensional models of each road are determined.

[0011] Based on the three-dimensional models of each road, determine the three-dimensional road features of each road;

[0012] Based on the three-dimensional road features of each road and the relationships between the roads, the road structure within the target area is obtained.

[0013] Preferably, determining the two-dimensional model of each road of the preset type within the target area based on the two-dimensional map includes:

[0014] Discretize the two-dimensional map to obtain discrete points;

[0015] Based on the discrete points, determine the road intersections, and based on the road intersections, determine the road discrete points corresponding to each road from the discrete points;

[0016] By fitting the discrete points corresponding to each road, a two-dimensional model of each road is obtained.

[0017] Preferably, based on the two-dimensional model of each road and the corresponding elevation data, a three-dimensional model of each road is determined, including:

[0018] Add corresponding elevation data to the discrete points of each road in the two-dimensional model of each road;

[0019] The elevation data added to the two-dimensional model of each road is fitted to obtain the three-dimensional model of each road.

[0020] Preferably, the three-dimensional road features of each road are determined based on the three-dimensional model of each road, including:

[0021] The radius of curvature, longitudinal slope, and length characteristics of each road are determined based on the three-dimensional model of the road.

[0022] Preferably, the road structure within the target area is obtained based on the three-dimensional road features of each road and the relationships between the roads, including:

[0023] Based on the three-dimensional road features of each road and the relationships between the roads, the road structure within the target area is obtained as follows:

[0024] Class = (Ra, Slo, Len)

[0025] where Ra∈{ra i ,i=1,2,...,n}

[0026] Slo∈{slo i ,i=1,2,...,n}

[0027] Len∈{len i ,i=1,2,...,n}

[0028] Where Ra, Slo, and Len represent the radius of curvature, longitudinal slope, and length characteristics of road segments with the same characteristic in the road structure, respectively, and n is the number of roads included in the road structure. i,slo i len i These represent the radius of curvature, longitudinal slope, and length characteristics of the corresponding feature segment in the i-th road.

[0029] Preferably, obtaining a two-dimensional map of roads of a preset type within the target area includes:

[0030] Obtain the two-dimensional map corresponding to the target area from OpenStreeMap;

[0031] Obtain the two-dimensional map corresponding to the road of the preset type from the two-dimensional map corresponding to the target area.

[0032] Preferably, the elevation data corresponding to roads of a preset type within the target area is obtained, including:

[0033] Obtain the digital elevation model of the target area from the NASA website;

[0034] Obtain the elevation data corresponding to the preset type of road from the digital elevation model.

[0035] A road structure recognition device, comprising:

[0036] The acquisition module is used to acquire two-dimensional maps and elevation data corresponding to roads of preset types within the target area;

[0037] The first determining module is used to determine the two-dimensional model of each road of the preset type in the target area based on the two-dimensional map;

[0038] The second determining module is used to determine the three-dimensional model of each road based on the two-dimensional model of each road and the corresponding elevation data.

[0039] The third determining module is used to determine the three-dimensional road features of each road based on the three-dimensional model of each road.

[0040] A road structure module is obtained, which is used to obtain the road structure within the target area based on the three-dimensional road features of each road and the relationship between the roads.

[0041] A road structure recognition device, comprising:

[0042] Memory, used to store computer programs;

[0043] A processor for executing the computer program to implement the steps of the road structure recognition method as described in any of the preceding claims.

[0044] A readable storage medium storing a computer program that, when executed by a processor, implements the steps of the road structure recognition method as described in any of the preceding claims.

[0045] This application provides a road structure identification method, apparatus, device, and readable storage medium. The method includes: acquiring a two-dimensional map and elevation data corresponding to roads of a preset type within a target area; determining a two-dimensional model of each road of the preset type within the target area based on the two-dimensional map; determining a three-dimensional model of each road based on the two-dimensional model and corresponding elevation data; determining the three-dimensional road features of each road based on the three-dimensional model; and obtaining the road structure within the target area based on the three-dimensional road features and the relationships between roads.

[0046] The technical solution disclosed in this application determines the two-dimensional model of each road based on the two-dimensional map corresponding to the preset type of road in the target area, and determines the three-dimensional model of each road in combination with the corresponding elevation data. Then, the three-dimensional road features of each road are determined based on the three-dimensional model of each road, and the road structure in the target area is obtained based on the three-dimensional road features of each road. Since the three-dimensional model contains spatial location information, it can more completely reflect the objective world and better fit the actual road conditions. Therefore, compared with the existing road structure recognition based on two-dimensional planar maps, the accuracy of road structure recognition based on three-dimensional roads in this application is relatively high. Therefore, it can better provide road constraints for network communication and resource allocation in the Internet of Vehicles. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0048] Figure 1 A flowchart illustrating a road structure identification method provided in this application embodiment;

[0049] Figure 2 This is a schematic diagram of a two-dimensional map corresponding to a preset type of road within the Fourth Ring Road of Beijing, provided in an embodiment of this application.

[0050] Figure 3 A schematic diagram of a three-dimensional model of certain roads identified in an embodiment of this application;

[0051] Figure 4 A schematic diagram of discrete points obtained after discretizing a two-dimensional map corresponding to a certain intersection in an embodiment of this application;

[0052] Figure 5 A schematic diagram of a road intersection provided in an embodiment of this application;

[0053] Figure 6 This is a schematic diagram of an unfitted two-dimensional road provided in an embodiment of this application;

[0054] Figure 7 A specific road structure identification diagram provided for an embodiment of this application;

[0055] Figure 8 This application provides a two-dimensional map of the area within the Fourth Ring Road of Beijing.

[0056] Figure 9 This is a schematic diagram of a road structure recognition device provided in an embodiment of this application;

[0057] Figure 10 This is a schematic diagram of a road structure recognition device provided in an embodiment of this application. Detailed Implementation

[0058] The future of fully autonomous driving requires sensing passenger intentions, vehicle conditions, and road traffic conditions, and then analyzing, predicting, and judging these factors to provide passengers with a safe and comfortable experience and to ensure the efficient operation of the transportation system. Due to limitations in computing power and perception range, the intelligence of a single vehicle cannot achieve this goal. Vehicle-to-everything (V2X) systems with edge intelligence can provide ultra-low latency information sharing and enhanced data analysis capabilities between vehicles, making fully autonomous driving possible.

[0059] Currently, existing vehicle-to-everything (V2X) edge computing is mainly based on two-dimensional plane maps for road structure identification. The identified road structure is then used to provide road constraints for network communication and resource allocation in the V2X. However, two-dimensional plane maps lose spatial location information and cannot fully and accurately reflect objective time. Therefore, the accuracy of road structure identification based on two-dimensional plane maps is not high.

[0060] Therefore, this application provides a road structure identification method, apparatus, device, and readable storage medium to improve the accuracy of road structure identification.

[0061] 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.

[0062] See Figure 1 The diagram illustrates a flowchart of a road structure identification method provided in an embodiment of this application. This road structure identification method may include:

[0063] S11: Obtain the two-dimensional map and elevation data corresponding to the preset type of roads within the target area.

[0064] When performing road structure recognition, the target area and the preset road type can be determined first, such as the area within the Fourth Ring Road of Beijing. The specific preset road type can be set as needed, such as highways, urban arterial roads, national highways, etc.

[0065] Then, a two-dimensional map and corresponding elevation data for roads of a preset type within the defined target area can be obtained. Specifically, the two-dimensional map mentioned here is a two-dimensional vector map, and the coordinates in this map are composed of the longitude and latitude of the corresponding location. The elevation data specifically refers to the elevation of a particular point, expressed in numerical form. See [link to documentation] for details. Figure 2 It shows a two-dimensional map diagram of preset types of roads within the Fourth Ring Road of Beijing provided in this application embodiment, wherein, Figure 2 The image shown is a two-dimensional map of Beijing's main roads. Figure 2 This is merely an example and does not limit the target area mentioned in this application to Beijing. Of course, it can also be other areas, and this application does not limit it.

[0066] S12: Determine the two-dimensional model of each road of the preset type within the target area based on the two-dimensional map.

[0067] Based on step S11, the two-dimensional models of each road in the preset type of roads within the target area can be determined according to the obtained two-dimensional map corresponding to the preset type of roads within the target area. That is, the two-dimensional models of each road in the preset type of roads within the target area are constructed using the obtained two-dimensional map, so as to construct the corresponding three-dimensional models of each road based on the two-dimensional models. Here, the two-dimensional model refers to the model formed by longitude and latitude.

[0068] S13: Based on the two-dimensional models of each road and the corresponding elevation data, determine the three-dimensional models of each road.

[0069] Based on the aforementioned steps, the three-dimensional models of each road can be determined according to the two-dimensional models of each road of a preset type within the target area and the corresponding elevation data. The three-dimensional model mentioned here is the model formed by adding elevation data to the two-dimensional model.

[0070] As the above process demonstrates, the 3D model of a road contains spatial location information, reflecting the objective world as completely as possible. It reduces the abstraction of the real world, focusing instead on a more intuitive description of it. For details, please refer to... Figure 3 It shows a schematic diagram of a three-dimensional model of certain roads identified in the embodiments of this application. It can be seen that the three-dimensional model simulation of roads takes realistic factors into account. Figure 3 All subsequent attached diagrams with coordinate systems use Cartesian coordinate systems. In the two-dimensional coordinate system, the x-axis and y-axis represent longitude and latitude, respectively. In the three-dimensional coordinate system, the x-axis, y-axis, and z-axis represent longitude, latitude, and elevation, respectively.

[0071] S14: Determine the three-dimensional road features of each road based on its three-dimensional model.

[0072] Based on step S13, the three-dimensional road features of each road can be determined using statistical and data correlation methods based on the three-dimensional model of each road.

[0073] S15: Based on the three-dimensional road characteristics of each road and the relationship between roads, the road structure within the target area is obtained.

[0074] After determining the three-dimensional road features of each road, the road structure within the target area can be obtained based on these features and the relationships between roads (which can be obtained from two-dimensional maps and three-dimensional models). Based on this, the road structure within each sub-region of the target area can then be determined.

[0075] In addition, when a vehicle is at any location within the target area, the three-dimensional road features and corresponding road structures of each road in the vicinity of that location can be obtained.

[0076] As can be seen from the above process, this application improves the accuracy of road structure recognition by establishing a three-dimensional model of the road, determining the three-dimensional road features based on the three-dimensional model, and obtaining the road structure based on the three-dimensional road features, thereby providing a better road foundation for the research on autonomous driving of vehicles.

[0077] The technical solution disclosed in this application determines the two-dimensional model of each road based on the two-dimensional map corresponding to the preset type of road in the target area, and determines the three-dimensional model of each road in combination with the corresponding elevation data. Then, the three-dimensional road features of each road are determined based on the three-dimensional model of each road, and the road structure in the target area is obtained based on the three-dimensional road features of each road. Since the three-dimensional model contains spatial location information, it can more completely reflect the objective world and better fit the actual road conditions. Therefore, compared with the existing road structure recognition based on two-dimensional planar maps, the accuracy of road structure recognition based on three-dimensional roads in this application is relatively high. Therefore, it can better provide road constraints for network communication and resource allocation in the Internet of Vehicles.

[0078] This application provides a road structure recognition method that determines the two-dimensional models of each road of a preset type within a target area based on a two-dimensional map. The method may include:

[0079] Discretize the two-dimensional map to obtain discrete points;

[0080] Determine road intersections based on discrete points, and then determine the corresponding discrete points for each road from the discrete points based on the road intersections.

[0081] By fitting the discrete points corresponding to each road, a two-dimensional model of each road is obtained.

[0082] In this application, considering that the data in the acquired two-dimensional map is vector data, and the curve is stored in the way of straight lines to represent curves, the deviation is relatively large. Therefore, when determining the two-dimensional model of each road in the preset type of road in the target area based on the two-dimensional map, this application can adopt the method of first discretizing and then fitting to obtain the two-dimensional model of each road, so as to improve the accuracy and precision of the two-dimensional model of the road, thereby facilitating the improvement of the accuracy of road structure recognition.

[0083] Specifically, the data in the two-dimensional map can be discretized according to a preset interval (e.g., 1m) to obtain multiple discrete points, as follows: Figure 4 As shown, it illustrates a schematic diagram of discrete points obtained after discretizing a two-dimensional map corresponding to a certain intersection provided in the embodiments of this application.

[0084] Since it is known that the roads between any two adjacent intersections are the same road, we can first find the road intersections on the 2D map to classify the roads and categorize the discrete points contained in each road. Specifically, after discretizing the 2D map to obtain discrete points, we can first use cluster analysis to find areas with dense discrete points (these areas are the road intersection areas). For example, we can use cluster analysis to cluster the obtained discrete points into two classes: one class is the road intersection class (i.e., the discrete points contained in this class are road intersections), and the other class is the road discrete points (i.e., discrete points other than road intersections). Then, we use vector analysis to determine the intersection center points and road intersections in the clusters. Given the road center points and road intersections, we classify all discrete points in this area according to the Euclidean distance between discrete points and the road intervals, thus obtaining the road discrete points corresponding to each road. Specifically, as follows... Figure 5 The diagram illustrates a road intersection according to an embodiment of this application. The data numbers represent road intersections, which are discrete points. Due to the large range, the point intervals are 1m, resulting in an overall linear appearance. Figure 5 In the diagram, the roads branching off at each number are distinct, with the two nearly straight roads in the middle being distinct roads as well. Between numbers 7 and 8, one side of the arc represents one road, and the other side represents a different road. Similarly, between numbers 5 and 12, one side of the arc represents one road, and the other side represents a different road. For a more intuitive understanding, refer to [reference needed]. Figure 6 It shows an unfitted two-dimensional road diagram provided in the embodiments of this application, wherein each road corresponds to a large number of road discrete points.

[0085] After obtaining the discrete points of each road, a linear fitting method (such as the least squares method) can be used to fit the discrete points of each road to obtain a two-dimensional model of each road.

[0086] The above process can improve the accuracy and precision of the two-dimensional road model, thereby facilitating the improvement of the accuracy of three-dimensional road feature and road structure identification.

[0087] This application provides a road structure identification method that determines the three-dimensional model of each road based on its two-dimensional model and corresponding elevation data. The method may include:

[0088] Add corresponding elevation data to the discrete points of each road in the two-dimensional model;

[0089] The elevation data added to the two-dimensional model of each road is fitted to obtain the three-dimensional model of each road.

[0090] Considering that the elevation data accuracy is only 15m, which has a large error and exhibits discontinuity in three dimensions, a method similar to that used for building two-dimensional road models can be adopted to build three-dimensional road models to improve the accuracy of the three-dimensional models. Specifically, when determining the three-dimensional models of each road based on their two-dimensional models and corresponding elevation data, firstly, the elevation data corresponding to the discrete points of the roads used to construct the two-dimensional models of each road can be obtained from the elevation data obtained in step S11 (the elevation data can be obtained based on the coordinates of the discrete points). Then, the corresponding elevation data is added to the discrete points of each road in the two-dimensional models to transform the road discrete points from two-dimensional coordinates to three-dimensional coordinates. Afterward, linear fitting can be used to fit the added elevation data in the two-dimensional models of each road, based on the premise that the roads do not interfere with each other spatially and meet the basic characteristics of roads (such as continuity, road curvature at intersections, and length meeting standards), to obtain the three-dimensional models of each road.

[0091] Before fitting the elevation data added to the two-dimensional model of each road, the added elevation data can be adjusted according to the three-dimensional relationship between the roads (such as overpasses, culverts, etc.) (specifically, the adjustment is based on the actual road specifications) to improve the accuracy of the elevation data, thereby further improving the accuracy of the three-dimensional model of the road.

[0092] In addition, after obtaining the three-dimensional models of each road, cluster analysis can be used to cluster the roads based on the premise that they do not interfere with each other spatially and meet the basic characteristics of the roads. This allows vehicles to determine the category of a road in advance when driving on the road, thereby providing better services for vehicle driving.

[0093] This application provides a road structure identification method that determines the three-dimensional road features of each road based on its three-dimensional model, and may include:

[0094] The radius of curvature, longitudinal slope, and length characteristics of each road are determined based on the three-dimensional model of the road.

[0095] In this application, the radius of curvature, longitudinal slope and length characteristics of each road can be determined based on the three-dimensional model of each road, so as to determine the road structure based on these three-dimensional road characteristics.

[0096] It should be noted that a road may have multiple feature segments. Specifically, feature segments are determined based on the three-dimensional road features mentioned above. Among them, at least one of the following features of adjacent feature segments is different: radius of curvature, longitudinal slope, and length.

[0097] This application provides a road structure identification method that, based on the three-dimensional road features of each road and the relationships between roads, obtains the road structure within a target area, and may include:

[0098] Based on the three-dimensional road features of each road and the relationships between roads, the road structure within the target area is obtained as follows:

[0099] Class = (Ra, Slo, Len)

[0100] where Ra∈{ra i ,i=1,2,...,n}

[0101] Slo∈{slo i ,i=1,2,...,n}

[0102] Len∈{len i ,i=1,2,...,n}

[0103] Where Ra, Slo, and Len represent the radius of curvature, longitudinal slope, and length characteristics of road segments with the same feature in the road structure, respectively, and n is the number of roads included in the road structure. i ,slo i len i These represent the radius of curvature, longitudinal slope, and length characteristics of the corresponding feature segment in the i-th road.

[0104] In this application, based on the fact that the three-dimensional road features specifically include the radius of curvature, longitudinal slope, and length features, the road structure within the target area obtained according to the three-dimensional road features of each road and the relationships between roads is as follows:

[0105] Class = (Ra, Slo, Len)

[0106] where Ra∈{ra i ,i=1,2,...,n}

[0107] Slo∈{slo i ,i=1,2,...,n}

[0108] The above is a general model of road structure (i.e., a universal model of road structure). Ra, Slo, and Len represent the radius of curvature, longitudinal slope, and length characteristics of road segments with the same characteristics in the road structure, respectively. n is the number of roads included in the road structure. i ,slo i len i These represent the radius of curvature, longitudinal slope, and length characteristics of the corresponding feature segment in the i-th road.

[0109] Based on the existing road structure definition of two-dimensional roads studied in the Internet of Vehicles, they can be basically divided into straight roads, curves and intersections. Considering the differences in road dimensions in three-dimensional scenarios, they can be divided into roads with slopes and roads without slopes. Their representation is on Slo in Class=(Ra,Slo,Len), while the two-dimensional structural features are represented on the feature values ​​of Ra and Len.

[0110] When i≥3, the road structure is of the intersection type;

[0111] When ra i When the distance is ∈[30m,1000m], the road segment is a curve;

[0112] When ra i When the length is greater than 1000m, the road section is a straight road;

[0113] When Slo∈[0.3%,8%], the road segment is a slope;

[0114] When Slo∈[0,0.3%], the road segment has no longitudinal slope.

[0115] When a vehicle is at any location within the target area, the 3D road features of each road in the vicinity of that location can be acquired. Specifically, see [link to documentation]. Figure 7 This document illustrates a specific road structure recognition diagram provided in an embodiment of this application. It shows that there are 4 Ra values, 4 Slo values, and 4 Len values. The specific data can be obtained from the three-dimensional model of the road. It can be determined that the road ahead is an intersection, the left and right lanes are straight roads, and the road ahead is a straight road followed by an uphill slope. This constrains the vehicle's trajectory and allows the vehicle to select relevant roads to change its speed and acceleration.

[0116] This application provides a road structure recognition method that obtains a two-dimensional map corresponding to a preset type of road within a target area, which may include:

[0117] Obtain the 2D map corresponding to the target area from OpenStreeMap;

[0118] Retrieve the 2D map corresponding to the road of the preset type from the 2D map corresponding to the target area.

[0119] In this application, when obtaining a two-dimensional map corresponding to roads of a preset type within the target area, the two-dimensional map corresponding to the target area can be obtained from OpenStreeMap (OSM, a public map). The data file format of this two-dimensional map is .osm, as detailed below. Figure 8As shown, it illustrates a two-dimensional map of the area within the Fourth Ring Road of Beijing provided in this embodiment. Then, data corresponding to roads other than those of a preset type can be deleted from the two-dimensional map of the target area, and noise reduction and optimization can be performed to obtain a two-dimensional map corresponding to the preset type of roads (specifically, this can be done using ArcMap processing software).

[0120] OpenStreeMap is an online map collaboration project aiming to create a world map that is free to edit and open to all. OSM is a free, open-source, and editable map service built collaboratively by the online community.

[0121] The above process allows for a simple and quick way to obtain a two-dimensional map of roads of a preset type within the target area.

[0122] This application provides a road structure identification method that obtains elevation data corresponding to a preset type of road within a target area, which may include:

[0123] Obtain the digital elevation model of the target area from the NASA website;

[0124] Obtain elevation data corresponding to roads of a preset type from the digital elevation model.

[0125] In this application, when obtaining elevation data corresponding to roads of a preset type within the target area, a Digital Elevation Model (DEM) of the target area can be obtained from the NASA website. This DEM is in .tif raster data format. A Digital Elevation Model is a physical ground model that represents ground elevation using an ordered array of numerical values. It is a branch of Digital Terrain Models (DTMs). A DTM represents the spatial distribution of actual terrain features in digital form; sometimes, the terrain feature point refers only to the elevation of a ground point, and this type of digital terrain is described as a DEM. After obtaining the DEM for the target area, ArcMap processing software can be used to extract the elevation data corresponding to roads of a preset type within the target area from the DEM.

[0126] The above process allows for a simple and quick acquisition of elevation data for roads of a preset type within the target area. Alternatively, a digital elevation model (DEM) for the target area can be obtained from OpenStreetMap, and then ArcMap software can be used to extract the elevation data for roads of the preset type from the DEM.

[0127] This application also provides a road structure recognition device, see [link to relevant documentation]. Figure 9 It shows a schematic diagram of a road structure recognition device provided in an embodiment of this application, which may include:

[0128] Module 91 is used to acquire two-dimensional maps and elevation data corresponding to roads of a preset type within the target area;

[0129] The first determining module 92 is used to determine the two-dimensional model of each road in the preset type of road within the target area based on the two-dimensional map;

[0130] The second determining module 93 is used to determine the three-dimensional model of each road based on the two-dimensional model of each road and the corresponding elevation data.

[0131] The third determining module 94 is used to determine the three-dimensional road features of each road based on the three-dimensional model of each road.

[0132] The road structure module 95 is obtained, which is used to obtain the road structure within the target area based on the three-dimensional road features of each road and the relationship between roads.

[0133] This application provides a road structure identification device, wherein the first determining module 92 may include:

[0134] Discrete cells are used to discretize two-dimensional maps to obtain discrete points;

[0135] The first determining unit is used to determine road intersections based on discrete points, and to determine the road discrete points corresponding to each road from the discrete points based on the road intersections.

[0136] The first fitting unit is used to fit the discrete points of each road to obtain a two-dimensional model of each road.

[0137] This application provides a road structure identification device, wherein the second determining module 93 may include:

[0138] Add cells to add corresponding elevation data to the discrete points of each road in the two-dimensional model;

[0139] The second fitting unit is used to fit the elevation data added to the two-dimensional model of each road to obtain the three-dimensional model of each road.

[0140] This application provides a road structure identification device, wherein the third determining module 94 may include:

[0141] The second determining unit is used to determine the radius of curvature, longitudinal slope, and length characteristics of each road based on the three-dimensional model of the road.

[0142] This application provides a road structure identification device, wherein the road structure module 95 may include:

[0143] The road structure elements are obtained and used to determine the road structure within the target area based on the three-dimensional road features of each road and the relationships between roads:

[0144] Class = (Ra, Slo, Len)

[0145] where Ra∈{ra i ,i=1,2,...,n}

[0146] Slo∈{slo i ,i=1,2,...,n}

[0147] Len∈{len i ,i=1,2,...,n}

[0148] Where Ra, Slo, and Len represent the radius of curvature, longitudinal slope, and length characteristics of road segments with the same feature in the road structure, respectively, and n is the number of roads included in the road structure. i ,slo i len i These represent the radius of curvature, longitudinal slope, and length characteristics of the corresponding feature segment in the i-th road.

[0149] This application provides a road structure identification device, wherein the acquisition module 91 may include:

[0150] The first acquisition unit is used to acquire the two-dimensional map corresponding to the target area from OpenStreeMap;

[0151] The second acquisition unit is used to acquire a two-dimensional map of a preset type of road from the two-dimensional map corresponding to the target area.

[0152] This application provides a road structure identification device, wherein the acquisition module 91 may include:

[0153] The third acquisition unit is used to acquire digital elevation models of the target area from the NASA website;

[0154] The fourth acquisition unit is used to acquire elevation data corresponding to roads of a preset type from the digital elevation model.

[0155] This application also provides a road structure recognition device, see [link to relevant documentation]. Figure 10 It shows a schematic diagram of a road structure recognition device provided in an embodiment of this application, which may include:

[0156] Memory 101 is used to store computer programs;

[0157] When processor 102 executes a computer program stored in memory 101, it can perform the following steps:

[0158] Obtain the 2D map and elevation data corresponding to the preset type of roads within the target area; determine the 2D model of each road in the preset type of roads within the target area based on the 2D map; determine the 3D model of each road based on the 2D model and corresponding elevation data; determine the 3D road features of each road based on the 3D model of each road; and obtain the road structure within the target area based on the 3D road features and the relationships between roads.

[0159] This application embodiment also provides a readable storage medium storing a computer program, which, when executed by a processor, can perform the following steps:

[0160] Obtain the 2D map and elevation data corresponding to the preset type of roads within the target area; determine the 2D model of each road in the preset type of roads within the target area based on the 2D map; determine the 3D model of each road based on the 2D model and corresponding elevation data; determine the 3D road features of each road based on the 3D model of each road; and obtain the road structure within the target area based on the 3D road features and the relationships between roads.

[0161] The readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0162] For a description of the relevant parts of the road structure recognition device, equipment and readable storage medium provided in this application, please refer to the detailed description of the corresponding parts in the road structure recognition method provided in the embodiments of this application, and will not be repeated here.

[0163] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that the elements inherent in a process, method, article, or apparatus that includes a list of elements are included. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. Additionally, portions of the technical solutions provided in the embodiments of this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration.

[0164] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A road structure identification method, characterized in that, include: Obtain the 2D map and elevation data corresponding to roads of preset types within the target area; Based on the two-dimensional map, determine the two-dimensional model of each road of the preset type within the target area; Based on the two-dimensional models of each road and the corresponding elevation data, the three-dimensional models of each road are determined. Based on the three-dimensional models of each road, determine the three-dimensional road features of each road; Based on the three-dimensional road features of each road and the relationships between the roads, the road structure within the target area is obtained; Correspondingly, based on the three-dimensional models of each road, the three-dimensional road features of each road are determined, including: The radius of curvature, longitudinal slope, and length characteristics of each road are determined based on the three-dimensional model of the road. Correspondingly, based on the three-dimensional road features of each road and the relationships between the roads, the road structure within the target area is obtained, including: Based on the three-dimensional road features of each road and the relationships between the roads, the road structure within the target area is obtained as follows: ； ； ； ； in, , , These refer to the radius of curvature, longitudinal slope, and length characteristics of road segments sharing the same feature within the road structure. This refers to the number of roads included in the road structure. , , The first The characteristics of the radius of curvature, longitudinal slope, and length of the corresponding characteristic road segments in the road.

2. The road structure identification method according to claim 1, characterized in that, Determining a two-dimensional model of each road of the preset type within the target area based on the two-dimensional map includes: Discretize the two-dimensional map to obtain discrete points; Based on the discrete points, determine the road intersections, and based on the road intersections, determine the road discrete points corresponding to each road from the discrete points; By fitting the discrete points corresponding to each road, a two-dimensional model of each road is obtained.

3. The road structure identification method according to claim 2, characterized in that, Based on the two-dimensional models and corresponding elevation data of each road, a three-dimensional model of each road is determined, including: Add corresponding elevation data to the discrete points of each road in the two-dimensional model of each road; The elevation data added to the two-dimensional model of each road is fitted to obtain the three-dimensional model of each road.

4. The road structure identification method according to claim 1, characterized in that, Obtain a 2D map of roads of a preset type within the target area, including: Obtain the two-dimensional map corresponding to the target area from OpenStreeMap; Obtain the two-dimensional map corresponding to the road of the preset type from the two-dimensional map corresponding to the target area.

5. The road structure identification method according to claim 1, characterized in that, Obtain elevation data for roads of a preset type within the target area, including: Obtain the digital elevation model of the target area from the NASA website; Obtain the elevation data corresponding to the preset type of road from the digital elevation model.

6. A road structure identification device, characterized in that, include: The acquisition module is used to acquire two-dimensional maps and elevation data corresponding to roads of preset types within the target area; The first determining module is used to determine the two-dimensional model of each road of the preset type in the target area based on the two-dimensional map; The second determining module is used to determine the three-dimensional model of each road based on the two-dimensional model of each road and the corresponding elevation data. The third determining module is used to determine the three-dimensional road features of each road based on the three-dimensional model of each road. A road structure module is obtained, which is used to obtain the road structure within the target area based on the three-dimensional road features of each road and the relationship between the roads; Correspondingly, based on the three-dimensional models of each road, the three-dimensional road features of each road are determined, including: The radius of curvature, longitudinal slope, and length characteristics of each road are determined based on the three-dimensional model of the road. Correspondingly, based on the three-dimensional road features of each road and the relationships between the roads, the road structure within the target area is obtained, including: Based on the three-dimensional road features of each road and the relationships between the roads, the road structure within the target area is obtained as follows: ； ； ； ； in, , , These refer to the radius of curvature, longitudinal slope, and length characteristics of road segments sharing the same feature within the road structure. This refers to the number of roads included in the road structure. , , The first The characteristics of the radius of curvature, longitudinal slope, and length of the corresponding characteristic road segments in the road.

7. A road structure recognition device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the road structure recognition method as described in any one of claims 1 to 5.

8. A readable storage medium, characterized in that, The readable storage medium stores a computer program that, when executed by a processor, implements the steps of the road structure recognition method as described in any one of claims 1 to 5.

Images