Generation method and device of digital elevation model

Abstract

The invention discloses a method and a device for generating a digital elevation model. The method is applied to a mine scene and comprises the steps that vector map information of a to-be-driven area of a vehicle is acquired, and the to-be-driven area comprises a plurality of elevation points; based on the vector map information, respectively determining topographic feature types corresponding to the plurality of elevation points; dividing the vector map corresponding to the to-be-driven area into at least one closed polygon based on the topographic feature types corresponding to the plurality of elevation points; performing elevation smoothing processing on the frame of the at least one closed polygon to obtain at least one smoothed closed polygon; scanning the at least one smoothed closed polygon, and filling the polygonal surface of the at least one closed polygon with a scanning line; and constructing a digital elevation model corresponding to the to-be-driven area based on the filled polygonal surface. The technical problem of low generation efficiency of the digital elevation model is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned driving simulation technology, and in particular to a method and device for generating a digital elevation model. Background Art

[0002] In the field of autonomous mining, high-precision terrain data is a core element of the simulation system. As a key representation of terrain data, the accuracy and simulation efficiency of the Digital Elevation Model (DEM) are crucial to the development of autonomous driving technology. A DEM describes topography by recording ground elevation information in three-dimensional space. It is typically used to simulate the terrain characteristics of the area in which a vehicle travels and to provide key data such as background point clouds and visible light information for on-board sensors.

[0003] In related technologies, common methods for generating digital elevation models include constructing digital elevation models based on point cloud data obtained by radar (lidar). However, since radar hardware and corresponding data processing software are usually expensive and require reliance on a real environment, it is impossible to quickly build a digital elevation model that meets the requirements of the simulation system, which limits its application in the rapid iteration of simulation systems and the construction of virtual scenes.

[0004] Currently, no effective solution has been proposed to the technical problem of low efficiency in generating digital elevation models. Summary of the Invention

[0005] The embodiments of the present invention provide a method and apparatus for generating a digital elevation model, so as to at least solve the technical problem of low generation efficiency of the digital elevation model.

[0006] According to one aspect of an embodiment of the present invention, a method for generating a digital elevation model is provided, comprising: obtaining vector map information of an area to be traveled by a vehicle, wherein the vector map information is used to represent terrain features of the area to be traveled, and the area to be traveled includes a plurality of elevation points; based on the vector map information, determining the types of terrain features corresponding to the plurality of elevation points; based on the types of terrain features corresponding to the plurality of elevation points, dividing the vector map corresponding to the area to be traveled into at least one closed polygon, wherein the types of terrain features corresponding to the elevation points in a closed polygon are consistent; performing elevation smoothing on the border of the at least one closed polygon to obtain at least one smoothed closed polygon; scanning the at least one smoothed closed polygon, and filling the polygonal surface of the at least one closed polygon with a scan line; and constructing a digital elevation model corresponding to the area to be traveled based on the filled polygonal surface.

[0007] Optionally, based on the vector map information, the terrain feature types corresponding to the multiple elevation points are determined respectively, including: parsing the vector map information, and determining the attribute information corresponding to the multiple elevation points respectively, wherein the attribute information is used to at least represent the location information and elevation information of the corresponding elevation points; based on the attribute information corresponding to the multiple elevation points, the terrain feature types corresponding to the multiple elevation points are determined respectively.

[0008] Optionally, based on the terrain feature types corresponding to multiple elevation points, the vector map corresponding to the area to be driven is divided into at least one closed polygon, including: based on the terrain feature types corresponding to multiple elevation points, identifying elevation points with the same terrain feature type from multiple elevation points; based on the position information of the elevation points with the same terrain feature type, dividing the elevation points with the same terrain feature type into at least one area; based on the boundary of at least one area, generating at least one closed polygon.

[0009] Optionally, elevation smoothing is performed on the border of at least one closed polygon to obtain at least one smoothed closed polygon, including: determining the distance between each elevation point on the border of at least one closed polygon and each adjacent elevation point within a preset range; assigning a distance weight to each adjacent elevation point based on the distance between each elevation point and each adjacent elevation point within the preset range, wherein the distance weight corresponding to the adjacent elevation point that is closer to each elevation point is greater; based on the elevation values ​​of each adjacent elevation point corresponding to each elevation point and the distance weight corresponding to each adjacent elevation point, the elevation value corresponding to each elevation point is adjusted to obtain a target elevation value corresponding to each elevation point; based on the target elevation value corresponding to each elevation point, elevation smoothing is performed on the border of at least one closed polygon to obtain at least one smoothed closed polygon.

[0010] Optionally, based on the elevation values ​​of each adjacent elevation point corresponding to each elevation point and the distance weights corresponding to each adjacent elevation point, the elevation value corresponding to each elevation point is adjusted to obtain the target elevation value corresponding to each elevation point, including: taking a weighted average of the elevation values ​​of each adjacent elevation point corresponding to each elevation point and the distance weights corresponding to each adjacent elevation point to obtain a weighted average value corresponding to each elevation point; and replacing the elevation value corresponding to each elevation point with the weighted average value corresponding to each elevation point to obtain the target elevation value corresponding to each elevation point.

[0011] Optionally, the at least one smoothed closed polygon is scanned to fill the polygonal surface of the at least one closed polygon, including: performing detail enhancement processing on the at least one smoothed closed polygon to obtain the at least one enhanced closed polygon; encrypting the border of the at least one enhanced closed polygon to obtain the at least one encrypted closed polygon; and scanning the at least one encrypted closed polygon to fill the polygonal surface of the at least one closed polygon.

[0012] Optionally, detail enhancement processing is performed on multiple elevation points in at least one smoothed closed polygon to obtain at least one processed closed polygon, including: determining the noise type corresponding to the at least one closed polygon based on the terrain feature types corresponding to the multiple elevation points in the at least one smoothed closed polygon; generating noise values ​​matching the multiple elevation points based on the noise type and the position information corresponding to the multiple elevation points; adding the noise values ​​to the elevation values ​​of the corresponding elevation points, enhancing the terrain features of the multiple elevation points, and obtaining at least one enhanced closed polygon.

[0013] Optionally, the border of the at least one enhanced closed polygon is encrypted to obtain the at least one encrypted closed polygon, including: determining an encryption interval for encrypting the border of the at least one enhanced closed polygon; inserting at least one new elevation point on each border of the at least one enhanced closed polygon based on the encryption interval by a linear interpolation technique; determining an elevation value corresponding to the inserted at least one new elevation point; adding the at least one new elevation point and the elevation value corresponding to the at least one new elevation point to the border of the at least one closed polygon to obtain the at least one encrypted closed polygon.

[0014] Optionally, scanning processing is performed on the at least one encrypted closed polygon to fill the polygonal surface of the at least one closed polygon, including: scanning the at least one encrypted closed polygon with a scan line to fill the polygonal surface of the at least one closed polygon.

[0015] Optionally, a scan line algorithm is used to scan at least one encrypted closed polygon to fill the polygonal surface of at least one closed polygon, including: a scanning step of controlling the scanning line to scan the at least one encrypted closed polygon in one direction; a determining step of determining the elevation value corresponding to the intersection point in response to the scanning line intersecting with the border of the at least one encrypted closed polygon; an adding step of adding the intersection point and the elevation value corresponding to the intersection point to the at least one encrypted closed polygon; and repeating the scanning step, the determining step and the adding step until the polygonal surface of the at least one encrypted closed polygon is completely filled.

[0016] Optionally, based on the filled polygonal surface, a digital elevation model corresponding to the area to be driven is constructed, including: constructing a spatial grid on the filled polygonal surface; assigning the elevation values ​​corresponding to the elevation points on the polygonal surface to the corresponding spatial grids to obtain the target elevation values ​​corresponding to each spatial grid; based on the target elevation values ​​corresponding to each spatial grid, a digital elevation model corresponding to the area to be driven is constructed.

[0017] Optionally, the elevation values ​​corresponding to the elevation points on the polygonal surface are assigned to the corresponding spatial grids to obtain the target elevation values ​​corresponding to each spatial grid, including: determining the elevation points corresponding to each spatial grid on the polygonal surface; determining the distance weights of the elevation points corresponding to each spatial grid based on the distances between the elevation points corresponding to each spatial grid and the center points of the corresponding spatial grids; and performing weighted average calculation of the elevation values ​​of the elevation points corresponding to each spatial grid based on the distance weights of the elevation points corresponding to each spatial grid to obtain the target elevation values ​​corresponding to each spatial grid.

[0018] Optionally, the method further includes: establishing a spatial index for each spatial grid on the polygonal surface, wherein the spatial index is used to query a target elevation value corresponding to the spatial grid.

[0019] According to another aspect of an embodiment of the present invention, a device for generating a digital elevation model is also provided, including: an acquisition unit for acquiring vector map information of an area to be traveled by a vehicle, wherein the vector map information is used to represent the terrain features of the area to be traveled, and the area to be traveled includes multiple elevation points; a parsing unit for determining the terrain feature types corresponding to the multiple elevation points based on the vector map information; a division unit for dividing the vector map corresponding to the area to be traveled into at least one closed polygon based on the terrain feature types corresponding to the multiple elevation points, wherein the terrain feature types corresponding to the elevation points in a closed polygon are consistent; a processing unit for performing elevation smoothing on the border of the at least one closed polygon to obtain at least one smoothed closed polygon; a scanning unit for scanning the at least one smoothed closed polygon and filling the polygonal surface of the at least one closed polygon through a scan line; and a filling unit for constructing a digital elevation model corresponding to the area to be traveled based on the filled polygonal surface.

[0020] According to another aspect of an embodiment of the present invention, an electronic device is provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods of various embodiments of the present invention when running.

[0021] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0022] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0023] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0024] According to another aspect of the embodiments of the present invention, a computer program is provided. When the computer program is executed by a processor, the methods in various embodiments of the present invention are implemented.

[0025] In an embodiment of the present invention, vector map information of the area to be traveled is directly used to determine the type of terrain feature corresponding to each elevation point in the area to be traveled. Based on the type of terrain feature corresponding to each elevation point in the area to be traveled, the vector map is divided into a series of closed polygons containing elevation points of the same terrain feature. This process does not require additional hardware equipment to collect data, significantly reducing initial costs. Then, by performing elevation smoothing on the borders of the closed polygons, abnormal mutations in the elevation data are eliminated, improving the visual and physical consistency of the subsequently constructed digital elevation model. Subsequently, efficient scan line technology is used to fill the polygonal surface to quickly construct a complete terrain surface. Moreover, the terrain features of the digital elevation model directly generated based on the vector map information are strictly consistent with the original map data, with rich terrain details while avoiding the uncertainty introduced by information mismatch. This greatly enhances the reliability and authenticity of the simulation system, thereby improving the accuracy and efficiency of the construction of the digital elevation model, thereby solving the technical problem of low generation efficiency of the digital elevation model. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0027] Figure 1 is a flow chart of a method for generating a digital elevation model according to an embodiment of the present invention;

[0028] Figure 2is a flow chart of another method for generating a digital elevation model according to an embodiment of the present invention;

[0029] Figure 3 4 is a schematic diagram of a device for generating a digital elevation model according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0031] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or apparatuses.

[0032] According to an embodiment of the present invention, an embodiment of a method for generating a digital elevation model is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0033] Figure 1 is a flow chart of a method for generating a digital elevation model according to an embodiment of the present invention, which is applied to a mining environment, such as Figure 1 As shown, the method includes the following steps:

[0034] Step S101: obtaining vector map information of the area where the vehicle is to travel.

[0035] In the technical solution provided in step S101 above, the area to be driven by the vehicle may be a mining area. The distances shown here are merely exemplary and do not specifically limit the area to be driven by the vehicle. The vector map information of the area to be driven is used to characterize the terrain characteristics of the area to be driven, and the area to be driven includes multiple elevation points, each of which reflects the surface height at the location.

[0036] In this embodiment, vector map information of the area to be traveled by the vehicle is obtained. The vector map information is digitized geographic information that can record the exact location and attributes of each geographic entity in the area to be traveled. For a mining environment, these entities may include various terrain elements such as roads, buildings, rivers, and vegetation. This is only an illustrative example.

[0037] Step S102: determining the terrain feature types corresponding to the plurality of elevation points based on the vector map information.

[0038] In the technical solution provided in the above-mentioned step S102 of the present application, it can be seen from the introduction of step S101 that the vector map information is used to characterize the terrain characteristics of the area to be driven. Based on this, by parsing the vector map information, the terrain feature types corresponding to multiple elevation points in the area to be driven can be obtained, wherein the terrain feature types are used to indicate different categories of landforms, such as roads, buildings, rivers, vegetation, etc. Each terrain feature type has its own unique elevation distribution pattern and morphological characteristics.

[0039] In this embodiment, by parsing the vector map information, the location information of each elevation point can be extracted from the vector map. This may involve extracting point data from multiple different types of geographic elements, such as road networks, mountain contours, and water edges. For each elevation point, the geographic element type is determined, thereby marking the corresponding terrain feature type. Geographic element types may include road element types, building element types, river element types, and vegetation element types.

[0040] For example, if an elevation point is located on a road feature, the terrain feature type of the elevation point is marked as “road.” According to this method, the terrain feature type corresponding to each elevation point among multiple elevation points can be determined.

[0041] Step S103 : dividing the vector map corresponding to the area to be driven into at least one closed polygon based on the terrain feature types corresponding to the plurality of elevation points.

[0042] In the technical solution provided in the above step S103 of the present application, after obtaining the terrain feature types corresponding to multiple elevation points through step S102, the elevation points with the same terrain feature type can be used to divide the vector map corresponding to the area to be traveled into at least one closed polygon.

[0043] In this embodiment, after determining the terrain feature types corresponding to the multiple elevation points, the vector map corresponding to the area to be driven is divided into at least one closed polygon. A closed polygon is a polygonal area with completely enclosed boundaries and no self-intersections. This division method facilitates subsequent operations such as boundary smoothing and interior filling, as closed polygons provide clear boundaries and facilitate independent processing of local areas.

[0044] Optionally, based on the terrain feature type, elevation points with the same terrain feature type are clustered together to form at least one closed polygon. For example, if the terrain feature type of multiple elevation points in the area to be traveled is the same, the vector map corresponding to the area to be traveled can be divided into a single closed polygon. If the terrain feature types of multiple elevation points in the area to be traveled are different, elevation points with the same terrain feature type can be clustered together to form multiple closed polygons.

[0045] For example, assuming that the terrain feature types corresponding to multiple elevation points include hill terrain type and river terrain type, the multiple elevation points belonging to the hill terrain type can be classified into one category, and the multiple elevation points belonging to the plain terrain type can be classified into one category. After determining the elevation points belonging to different terrain feature types, the elevation points in each terrain feature type can be divided into at least one closed polygon based on the position information of the elevation points in each terrain feature type. For example, if the elevation points in the same terrain feature type are not divided by the elevation points of other terrain feature types, the elevation points of the same terrain feature type can be divided into one closed polygon. If the elevation points of the same terrain feature type are divided by the elevation points of other terrain feature types, the elevation points of the same terrain feature type that are close to each other in the same segmentation area can be divided into multiple closed polygons based on the segmentation situation and the information of the elevation points of the same terrain feature type.

[0046] In this step, the vector map corresponding to the area to be driven is divided based on the terrain feature types corresponding to multiple elevation points. This not only simplifies the complex terrain features into easy-to-process polygonal units, but also ensures that the terrain description within each unit is relatively uniform, which facilitates subsequent elevation information processing and digital elevation model construction.

[0047] Step S104: performing elevation smoothing on the border of at least one closed polygon to obtain at least one smoothed closed polygon.

[0048] In the technical solution provided in the above step S104 of the present application, after the vector map of the driving area is divided into at least one closed polygon, the elevation points on the closed polygon border may have large local elevation differences due to data collection errors or other factors, forming a "spike" or "cliff" effect. If these abnormal elevation points are not processed, they will appear as abrupt landform changes in the final digital elevation model, which will not only reduce the realism of the terrain, but may also affect the accuracy and stability of the unmanned driving system simulation. Therefore, the closed polygon border after division can be smoothed in elevation to eliminate abnormal mutations in the terrain and ensure the continuity and rationality of the terrain surface.

[0049] In this embodiment, the elevation point data on the closed polygonal border can be read. These elevation point data constitute the outline of the polygonal border and are also the objects of subsequent smoothing processing. Afterwards, the elevation points on the border are smoothed using a distance weighting method.

[0050] For example, taking an elevation point on a closed polygonal border as an example, for the sake of convenience, this elevation point can be called a target elevation point. With the target elevation point as the center point, select elevation points on the closed polygonal border that are within a preset range from the target elevation point. Then, according to the distance between the selected elevation point and the target elevation point, set the distance weight of each elevation point. The closer the elevation point is to the target elevation point, the greater the distance weight, that is, the greater the impact on the target elevation point. After determining the distance weight corresponding to each elevation point, the elevation values ​​of the selected elevation points can be weighted averaged according to the weight value corresponding to each elevation point to obtain the target elevation value, and the target elevation value can be used to update the elevation value of the target elevation point. Referring to this method, after updating the elevation values ​​of multiple elevation points on the closed polygonal border, the closed polygon obtained is a closed polygon after elevation smoothing.

[0051] In this step, by performing elevation smoothing on the border of the closed polygon, abnormal elevation points in the closed polygon can be eliminated, so that the final digital elevation model does not include abrupt landform changes.

[0052] Step S105 : Scan the at least one smoothed closed polygon, and fill the polygonal surface of the at least one closed polygon with the scan lines.

[0053] In the technical solution provided in step S105 of the present application, after performing elevation smoothing on at least one closed polygon in step S104, to further reduce holes in the polygon, the closed polygon may be densified using a linear interpolation method. Densification refers to inserting new calculation points between existing elevation points to make their distribution more dense.

[0054] In this embodiment, the scan line algorithm moves a virtual "scan line" along a specified direction (usually horizontally) inside a closed polygon, detects the intersection of this scan line and the polygon boundary line each time it moves, and then calculates and fills in the corresponding terrain elevation information based on these intersections.

[0055] For example, determine the starting and ending positions of the scan line, as well as the direction of movement. Typically, start from the lowest point and move upward along the positive Y-axis (vertical direction). As the scan line moves, the intersections of the scan line and the boundaries of each closed polygon are continuously detected. These intersections represent the locations where the terrain changes. For each detected intersection, calculate the elevation information of the closed polygon where the intersection is located. Use the calculated elevation values ​​to fill the terrain area between the intersections to form a continuous polygonal surface.

[0056] Optionally, in the process of filling the polygonal surface of the closed polygon with the scan line, appropriate elevation noise may be added to each elevation point to simulate subtle changes in the natural terrain, thereby enhancing the realism of the digital elevation model.

[0057] In this step, by filling the polygonal surface of at least one closed polygon with the scan line and detecting the intersection of the scan line and the closed polygon and the filled terrain information, the abstract closed polygon can be converted into a visual, continuous polygonal surface, laying a solid foundation for the final construction of the digital elevation model.

[0058] Step S106: constructing a digital elevation model corresponding to the area to be driven based on the filled polygonal surface.

[0059] In the technical solution provided in the above step S106 of the present application, after the filled polygonal surface is obtained, a digital elevation model corresponding to the area to be traveled can be constructed based on the filled polygonal surface.

[0060] In this embodiment, a spatial grid covering the area to be driven can be constructed. This spatial grid consists of a series of regularly arranged nodes, forming a three-dimensional coordinate framework. The size and density of the grid can be adjusted based on the required accuracy and complexity of the digital elevation model, ensuring that the final model accurately reflects the terrain features without excessively increasing the computational load.

[0061] Optionally, a spatial index is created for the filled polygons to efficiently search and match elevation information. A spatial index is a data structure that allows for rapid location and retrieval of elevation data at specific locations. The spatial index is then used to calculate the elevation value of each node in the spatial grid using a nearest neighbor matching algorithm. Nearest neighbor matching involves finding the closest polygon filling point or points to the current grid node and assigning the elevation value of those filling points to the grid node. This process ensures the continuity and accuracy of the elevation information in the digital elevation model.

[0062] Optionally, to ensure a smooth transition and natural appearance of the digital elevation model, the calculated spatial grid elevation values ​​are further smoothed. This can be achieved using a moving average, Gaussian filtering, or other algorithms to remove any possible elevation anomalies and make the terrain surface appear more natural. Depending on the actual simulation requirements, noise information can be optionally added to the grid nodes to simulate subtle variations in the actual terrain, such as height differences caused by rocks, potholes, or vegetation, thus making the digital elevation model more realistic and detailed.

[0063] Optionally, the processed spatial grid data is encoded and stored in a standard digital elevation model format to form a complete digital elevation model. This model not only contains accurate terrain elevation data, but also includes terrain feature types and detail enhancement information, which can fully reflect the terrain characteristics of the area to be driven.

[0064] Optionally, after obtaining the digital elevation model corresponding to the area to be driven, it can be applied to the simulation test of the unmanned vehicle to provide accurate terrain information and background environment data for the vehicle's dynamic model and sensor model.

[0065] In this step, the filled polygonal surface is efficiently converted into a digital elevation model through spatial grid construction, spatial index establishment, nearest neighbor matching calculation, elevation smoothing, and detail enhancement. This process not only ensures the accuracy and detail of the digital elevation model, but also greatly improves its generation efficiency, reduces human intervention and hardware dependence, and provides powerful and realistic terrain information support for the simulation of unmanned mining vehicles, significantly improving the reliability and effectiveness of simulation tests.

[0066] In steps S101 to S106, the vector map information of the area to be traveled is directly used to determine the type of terrain feature corresponding to each elevation point in the area to be traveled. Based on the type of terrain feature corresponding to each elevation point in the area to be traveled, the vector map is accurately divided into a series of closed polygons containing elevation points of the same terrain feature. This process does not require additional hardware equipment to collect data, significantly reducing initial costs. Next, by performing elevation smoothing on the borders of the closed polygons, abnormal mutations in the elevation data are eliminated to improve the visual and physical consistency of the subsequently constructed digital elevation model. Subsequently, efficient scan line technology is used to fill the polygon faces to quickly construct a complete terrain surface. Moreover, the digital elevation model directly generated based on the vector map information has terrain features that are strictly consistent with the original map data, possessing rich terrain details while avoiding the uncertainty introduced by information mismatches. This greatly enhances the reliability and authenticity of the simulation system, thereby improving the accuracy and efficiency of the digital elevation model construction, thereby solving the technical problem of low digital elevation model generation efficiency.

[0067] Next, the generation method of the digital elevation model is further introduced.

[0068] As an optional implementation, step S102, based on the vector map information, determines the terrain feature types corresponding to multiple elevation points respectively, including: parsing the vector map information, determining the attribute information corresponding to the multiple elevation points respectively, wherein the attribute information is used to at least represent the location information and elevation information of the corresponding elevation points; based on the attribute information corresponding to the multiple elevation points, determining the terrain feature types corresponding to the multiple elevation points respectively.

[0069] In this embodiment, vector map data typically includes the location and attribute information of a large number of geographic features, such as roads, hills, and buildings. These features are represented in vector format using points, lines, and surfaces, and are accompanied by rich attribute descriptions, such as elevation, material, and purpose. By reading and parsing this vector map data, it is possible to extract the elevation point information and attributes related to the terrain features.

[0070] Optionally, use a vector information reading tool to read the vector map file and extract the location information (i.e., latitude and longitude or coordinates) and elevation information from the attribute information of each elevation point. Other descriptions in the attribute information, such as terrain type, vegetation cover, and geological characteristics, can also be parsed to assist in determining the specific terrain feature type of the elevation point.

[0071] Optionally, after parsing the attribute information of the elevation points, the type of terrain feature corresponding to each elevation point can be determined based on the parsed information. For example, by comparing the elevation values ​​and location information of different elevation points, clusters of elevation points with similar terrain features can be identified. Based on the elevation value, change trend, and related attribute information (such as terrain type description) of the elevation point, the specific terrain feature type to which each elevation point belongs is determined. For example, if consecutive elevation points show a clear upward or downward trend and are marked as hills or valleys in the attribute information, then these points belong to the hill or valley terrain type.

[0072] In this step, by parsing the vector map information, elevation point data with detailed terrain feature type identification is obtained, which provides the necessary information support for the subsequent polygon division and construction of the digital elevation model.

[0073] As an optional implementation, step S103 divides the vector map corresponding to the area to be driven into at least one closed polygon based on the terrain feature types corresponding to multiple elevation points, including: identifying elevation points with the same terrain feature type from multiple elevation points based on the terrain feature types corresponding to multiple elevation points; dividing the elevation points with the same terrain feature type into at least one area based on the location information of the elevation points with the same terrain feature type; and generating at least one closed polygon based on the boundary of at least one area.

[0074] In this embodiment, the vector map information of the area to be traveled is divided into at least one closed polygon according to the terrain feature type of the elevation point, so that the area to be traveled is classified according to the terrain feature type.

[0075] Optionally, based on the vector map information and the terrain feature types of the multiple elevation points determined above, elevation points with the same terrain feature type are identified. For example, by analyzing and comparing the attribute information of the elevation points, elevation points with the same terrain features are grouped together. For example, all elevation points representing hilly terrain are separated and processed independently from elevation points representing plain terrain. This identification process utilizes additional information about the elevation points, such as geological characteristics, vegetation coverage, soil type, etc., making the model more realistic than a simple classification based on elevation values.

[0076] Optionally, after identifying elevation points with the same terrain characteristics, these elevation points with the same terrain characteristics can be divided into at least one region based on their location information. This division is achieved by connecting adjacent elevation points to form a closed polygonal boundary. The polygonal boundary should be completely closed and non-self-intersecting, ensuring that each generated polygonal region is valid and independent.

[0077] Optionally, if the elevation points of the same terrain feature type are not separated by elevation points of other terrain feature types, the elevation points of the same terrain feature type can be divided into a closed polygon. For example, the elevation points at the boundary of the elevation points of the same terrain feature type are connected to obtain a closed polygon. On the contrary, if the elevation points of the same terrain feature type are separated by elevation points of other feature types, it means that these elevation points are scattered in different areas. In this case, multiple closed polygons can be generated based on the position information of the elevation points of the same separated terrain feature type, wherein the terrain feature types corresponding to the elevation points included in each closed polygon are consistent. According to this method, the elevation points of the same terrain feature type can be divided into regions to obtain one or more closed polygons.

[0078] In this step, the division of closed polygonal areas takes into account the homogeneity of terrain feature types. By dividing the elevation points of the same terrain features into at least one closed polygonal area, the subsequent processing processes, such as elevation smoothing and polygon surface calculation, can be greatly simplified, avoiding unnecessary complex interactions between different terrain features and improving the construction efficiency of the entire digital elevation model.

[0079] As an optional implementation, step S104 performs elevation smoothing on the border of at least one closed polygon to obtain at least one smoothed closed polygon, including: determining the distance between each elevation point on the border of at least one closed polygon and each adjacent elevation point within a preset range; assigning a distance weight to each adjacent elevation point based on the distance between each elevation point and each adjacent elevation point within the preset range, wherein the distance weight corresponding to the adjacent elevation point that is closer to each elevation point is greater; adjusting the elevation value corresponding to each elevation point based on the elevation values ​​of each adjacent elevation point corresponding to each elevation point and the distance weight corresponding to each adjacent elevation point to obtain a target elevation value corresponding to each elevation point; performing elevation smoothing on the border of at least one closed polygon based on the target elevation value corresponding to each elevation point to obtain at least one smoothed closed polygon.

[0080] In this embodiment, when performing elevation smoothing on the border of at least one closed polygon, for each elevation point on the closed polygon border, a target elevation value can be obtained by performing weighted averaging on the elevation values ​​of multiple elevation points within a preset range of each elevation point, and then the elevation value of the corresponding elevation point is updated using the target elevation value. After the elevation value of each elevation point on the closed polygon border is updated, the closed polygon obtained is the closed polygon after smoothing.

[0081] Optionally, for each elevation point on the border of at least one closed polygon, the Euclidean distance between it and each neighboring elevation point within a preset range can be calculated. The "preset range" here refers to a reasonable radius set to limit computing resources, and only neighboring elevation points within this radius are considered. This approach avoids global calculations, improving processing speed and efficiency. Here, an arbitrary elevation point on the border of a closed polygon is used as an example. For ease of explanation, this elevation point is referred to as the target elevation point. Based on the distance between the target elevation point and its neighboring elevation points, a distance weight is assigned to each neighboring elevation point. The weight reflects the degree of influence of the neighboring elevation point on the elevation value of the target elevation point. For example, the closer the neighboring elevation point is to the target elevation point, the greater its elevation value influences the target elevation point, and the corresponding distance weight is larger. Conversely, the influence is smaller, and the distance weight is also smaller. Distance weights are introduced to ensure smooth transitions on the terrain surface and avoid unreasonable interference between the elevation values ​​of nearby points caused by distant points.

[0082] Optionally, based on the elevation values ​​of each neighboring elevation point and its corresponding distance weight, the original elevation value of the target elevation point can be adjusted to obtain an updated elevation value. After processing multiple elevation points on the closed polygon border separately according to this method, the obtained closed polygon is a smoothed closed polygon.

[0083] As an optional implementation, based on the elevation values ​​of each adjacent elevation point corresponding to each elevation point and the distance weights corresponding to each adjacent elevation point, the elevation value corresponding to each elevation point is adjusted to obtain the target elevation value corresponding to each elevation point, including: taking a weighted average of the elevation values ​​of each adjacent elevation point corresponding to each elevation point and the distance weights corresponding to each adjacent elevation point to obtain a weighted average value corresponding to each elevation point; and replacing the elevation value corresponding to each elevation point with the weighted average value corresponding to each elevation point to obtain the target elevation value corresponding to each elevation point.

[0084] In an embodiment, when adjusting the elevation value of each elevation point based on the information of the neighboring elevation points and the distance weight of each elevation point, it can usually be achieved by weighted averaging.

[0085] For example, taking the target elevation point as an example, the elevation value of each adjacent elevation point corresponding to the target elevation point can be multiplied by the distance weight and then summed, and finally divided by the sum of the weights of all adjacent elevation points to obtain the weighted average value corresponding to the target elevation point. Afterwards, the weighted average value of the target elevation point is used to replace the elevation value corresponding to the target elevation point (original elevation value) to obtain the target elevation value corresponding to the target elevation point. Referring to the adjustment method of the elevation value of the target elevation point, the elevation values ​​of multiple elevation points on the border of the closed polygon are updated, and the closed polygon border obtained is the closed polygon border after smoothing. The polygon border processed in this way has a more natural transition between elevation points, effectively eliminates possible elevation mutations, and is conducive to improving the quality and visual effect of the digital elevation model corresponding to the area to be driven.

[0086] In this step, the introduction of distance weights between neighboring elevation points ensures that elevation adjustments fully account for the local characteristics of the terrain, avoiding the distortion that might result from global averaging. Furthermore, the weighted average calculation method effectively balances the continuity and accuracy of terrain changes, providing smoother elevation information for subsequent polygon calculations.

[0087] As an optional implementation, step S105 scans at least one smoothed closed polygon to fill the polygonal surface of at least one closed polygon, including: performing detail enhancement processing on at least one smoothed closed polygon to obtain at least one enhanced closed polygon; encrypting the border of at least one enhanced closed polygon to obtain at least one encrypted closed polygon; scanning at least one encrypted closed polygon to fill the polygonal surface of at least one closed polygon.

[0088] In this embodiment, smoothing the bounding box of a closed polygon eliminates abnormal elevation data within the bounding box, ensuring the continuity and rationality of the terrain surface. However, smoothing often makes the terrain appear too flat, lacking the complexity and detail of natural terrain. For autonomous driving simulations in mining environments, realistic terrain features directly impact the accuracy of vehicle dynamics and sensor models. Therefore, detail enhancement is required on the smoothed closed polygon to enhance the realism of the terrain features.

[0089] Optionally, after performing detail enhancement processing on closed polygons, the border of at least one enhanced closed polygon can be encrypted to obtain at least one encrypted closed polygon, wherein encryption can reduce the holes in the polygon surface, ensure that as many terrain details as possible can be captured when the scan line is filled, and further improve the resolution and detail description capabilities of the closed polygon area.

[0090] For example, linear interpolation or another suitable algorithm can be used to increase the number of elevation points on a closed polygon's border, thereby densifying the border. For example, for each elevation point on the closed polygon's border, an interpolation calculation is performed based on the elevation information of its adjacent points to generate a new elevation point. This densified border is denser, providing more starting and ending points for subsequent polygon calculations.

[0091] Optionally, after encrypting the closed polygon, the encrypted closed polygon can be scanned to further fill in the polygonal surface of the closed polygon. For example, a continuous terrain surface can be constructed by moving a virtual scan line within the closed polygon, identifying intersections between the scan line and the boundary line, and then filling in the elevation information between these intersections.

[0092] For example, a horizontal scan line is chosen as the direction, starting at the lowest elevation point. Then, as the scan line moves in 3D space, its intersections with closed polygonal bounding boxes are dynamically detected. For each intersection, the elevation information of the nearest elevation points is interpolated to determine the elevation at that point. Finally, these calculated elevation values ​​are used to fill the polygonal surface, forming a complete, continuous terrain surface.

[0093] In this step, by enhancing the closed polygons' borders with details, encrypting them, and filling them with scan lines, a highly detailed and realistic polygonal surface is generated from the smoothed closed polygons, thereby improving the quality of the digital elevation model. This high-quality digital elevation model better reflects the actual conditions of the mining environment, providing a more reliable and realistic physical environment for unmanned vehicle simulation testing, thereby increasing the credibility and applicability of the simulation results.

[0094] As an optional implementation, detail enhancement processing is performed on multiple elevation points in at least one smoothed closed polygon to obtain at least one processed closed polygon, including: determining the noise type corresponding to the at least one closed polygon based on the terrain feature types corresponding to the multiple elevation points in the at least one smoothed closed polygon; generating noise values ​​matching the multiple elevation points based on the noise type and the position information corresponding to the multiple elevation points; adding the noise values ​​to the elevation values ​​of the corresponding elevation points, enhancing the terrain features of the multiple elevation points, and obtaining at least one enhanced closed polygon.

[0095] In this embodiment, when performing detail enhancement on multiple elevation points within at least one smoothed closed polygon, an appropriate noise type can be determined based on the terrain feature types corresponding to the multiple elevation points within the closed polygon. The noise type is selected based on the terrain feature type. For example, in hilly areas, noise can simulate the irregularities of rock distribution or vegetation cover; whereas in plain areas, the noise may be more subtle to represent slight ground undulations. The noise type is determined based on an understanding and study of real-world terrain characteristics, striving to ensure that the added noise appropriately reflects the natural variations of a specific terrain.

[0096] Optionally, after determining the noise type corresponding to each elevation point, a noise value can be generated to match it based on the selected noise type and the location information of each elevation point. This process typically involves generating random numbers, combining the terrain feature type and location information, to calculate a noise value appropriate for each elevation point. For example, a point at the top of a hill may require a larger noise value to simulate the distribution of rocks or vegetation, while a point at the foot of the hill may require a smaller noise value to maintain a smooth transition in the terrain.

[0097] Optionally, after determining the noise value corresponding to each elevation point, the calculated noise value can be added to the elevation value of the corresponding elevation point to enhance the closed polygon, so that the originally relatively smooth terrain surface obtains additional details, thereby simulating the real terrain characteristics of the area to be driven.

[0098] In this step, detail enhancement processing based on terrain feature types can not only compensate for the loss of terrain details caused by smoothing, but also further simulate and reproduce the characteristics of complex terrain and landforms, providing high-precision and high-realism terrain information for unmanned driving simulation in mining environments, greatly improving the quality and results of simulation tests.

[0099] As an optional implementation, the border of at least one enhanced closed polygon is encrypted to obtain the at least one encrypted closed polygon, including: determining an encryption interval for encrypting the border of at least one enhanced closed polygon; inserting at least one new elevation point on each border of the at least one enhanced closed polygon based on the encryption interval by a linear interpolation technique; determining an elevation value corresponding to the inserted at least one new elevation point; adding the at least one new elevation point and the elevation value corresponding to the at least one new elevation point to the border of the at least one closed polygon to obtain the at least one encrypted closed polygon.

[0100] In this embodiment, by encrypting the closed polygons after detail enhancement, the number of elevation points on the polygon border can be increased, making the final digital elevation model smoother and more realistic, which is especially important for the accurate reproduction of complex terrain in unmanned driving simulation systems in mines.

[0101] In this embodiment, the infill interval determines the density of newly added elevation points on the infilled border. The infill interval should be determined based on the resolution of the desired digital elevation model and the complexity of the terrain. For areas with drastic terrain changes or requiring high-precision modeling, a smaller infill interval should be set to increase the density of elevation points and ensure that terrain details are captured.

[0102] Optionally, after determining the encryption interval, new elevation points can be inserted on each polygonal border based on the encryption interval through linear interpolation technology. Among them, the linear interpolation method is used to determine the new elevation points inserted between two known elevation points. For example, assuming that there are two known elevation points A and B on the closed polygonal border, and the encryption interval is d, then a new elevation point C is inserted between A and B every d distance, and the elevation value of point C is calculated by linear interpolation. This ensures that the elevation value of the new point will not deviate from the elevation characteristics of the original point, and can fill the gaps in the elevation data on the border, making the border more continuous and dense.

[0103] Optionally, after inserting new elevation points, you need to determine the elevation of each new point. The elevation of each new point is a linear combination of the elevations of its two adjacent points, with the weight of the new point's elevation being inversely proportional to the distance from the two adjacent points to the new point. This calculated elevation ensures a smooth transition between the original point and the new one, avoiding sudden changes in the terrain model caused by the addition of new points.

[0104] Optionally, all new elevation points and their corresponding elevation values ​​are added to the original closed polygon border to form an encrypted closed polygon. This not only increases the number of elevation points on the closed polygon, but more importantly, by increasing the number and details of the elevation points, the resolution of the constructed digital elevation model and the accuracy of terrain details can be significantly improved.

[0105] In this step, the encrypted closed polygonal bounding boxes better reflect terrain variations, reducing voids in the model. This allows the resulting digital elevation model to more accurately reflect the topographical characteristics of the mining environment, enhancing the credibility and practical value of the unmanned driving simulation system. Furthermore, encryption facilitates subsequent polygon filling and digital elevation model construction, making the entire process more efficient and accurate, providing more accurate and comprehensive terrain data support for path planning and dynamic response simulation of unmanned mining vehicles.

[0106] As an optional implementation, scanning processing is performed on at least one encrypted closed polygon to fill the polygonal surface of at least one closed polygon, including: scanning at least one encrypted closed polygon using a scanning line to fill the polygonal surface of at least one closed polygon.

[0107] In this step, the at least one encrypted polygon may be scanned using a scan line algorithm to fill the space inside the polygon and construct a complete polygonal surface.

[0108] Alternatively, a scanline algorithm "sweeps" a closed area in a certain direction (usually horizontally) in two-dimensional or three-dimensional space, records the intersection point each time it encounters a boundary, and fills the area between the intersection points with a color or texture. By applying this scanline algorithm to polygon filling, the polygonal surface of a closed polygon is filled, that is, each point inside it is assigned an elevation value.

[0109] For example, a virtual scan line is defined, usually starting from the lowest point of the polygon and gradually moving upward (along the positive Y axis). Each movement of the scan line detects the intersection with the polygon boundary. When the scan line intersects the polygon boundary, the coordinate position of the intersection is calculated, including the X, Y, and Z coordinates (elevation values). Using the elevation points on both sides of the intersection, through linear interpolation or other methods, the exact elevation value of the intersection is calculated, ensuring the continuity and rationality of the elevation values ​​on the polygon surface.

[0110] Optionally, the elevation values ​​of the intersection points are used to fill in the area covered by the scan line, forming a segment of the polygon from one intersection point to the next. This process is repeated until the entire closed polygon is completely covered by the scan line, forming a complete polygon with elevation information.

[0111] In this step, the scanline algorithm fills closed polygons, meticulously constructing the internal terrain of the closed polygons. This seamlessly connects the polygon's elevation information with its border information, ensuring the continuity and smoothness of the terrain surface. This process not only greatly improves the efficiency of building digital elevation models but also ensures the realism of the terrain, providing reliable simulated terrain data for unmanned vehicles in mining environments.

[0112] As an optional implementation, a scan line algorithm is used to scan at least one encrypted closed polygon to fill the polygonal surface of at least one closed polygon, including: a scanning step of controlling the scanning line to scan the at least one encrypted closed polygon in one direction; a determining step of determining the elevation value corresponding to the intersection point in response to the scanning line intersecting with the border of the at least one encrypted closed polygon; an adding step of adding the intersection point and the elevation value corresponding to the intersection point to the at least one encrypted closed polygon; and repeating the scanning step, the determining step and the adding step until the polygonal surface of the at least one encrypted closed polygon is completely filled.

[0113] In this embodiment, the encrypted closed polygon is scanned using a scan line algorithm to fill the blank area inside the closed polygon and generate a polygonal surface, thereby constructing a more complete and realistic digital elevation model.

[0114] Optionally, define a virtual scan line that moves along a preset direction (usually perpendicular to the ground, i.e., the Y or Z axis) to scan the entire closed polygon area. The starting point of the scan line is usually located at the lowest elevation point of the polygon.

[0115] For example, a scan line starts from a starting point and gradually moves in a set direction until it passes through the entire closed polygon. During the entire movement process, the scan line will intersect the polygon's border line multiple times, and each intersection will trigger the determination and addition steps.

[0116] Optionally, when the scan line intersects the border line of the encrypted closed polygon, the specific position of the intersection, i.e., the X, Y coordinates (or X, Z coordinates, depending on the direction of the scan line), can be calculated. After the intersection is determined, the elevation value corresponding to the intersection can be calculated. Since the points on the closed polygon border have been encrypted and have relatively dense elevation information, the elevation value of the intersection can be obtained by linear interpolation of the nearest elevation points on both sides to ensure the continuity and rationality of the terrain surface. The calculated intersection position and its elevation value are recorded, and this information will be used for the next step of polygon filling.

[0117] Optionally, to ensure that the entire closed polygonal area can be scanned and filled, the scanning, determining and adding steps are repeated until the scan line completes coverage of the entire polygonal area, and the intersection points and their elevation values ​​are recorded completely, thereby forming a complete polygonal surface.

[0118] In this step, a scanline algorithm is used to effectively scan and fill the encrypted closed polygons. This process not only increases the detail of the digital elevation model but also ensures the continuity and accuracy of the terrain surface. For unmanned driving simulation systems in mining environments, this high-precision digital elevation model provides more realistic terrain information, helping vehicle models better understand and adapt to complex terrain conditions, thereby improving the credibility and effectiveness of simulation tests.

[0119] As an optional implementation, step S106 constructs a digital elevation model corresponding to the area to be driven based on the filled polygonal surface, including: constructing a spatial grid on the filled polygonal surface; assigning the elevation values ​​corresponding to the elevation points on the polygonal surface to the corresponding spatial grids to obtain the target elevation values ​​corresponding to each spatial grid; and constructing a digital elevation model corresponding to the area to be driven based on the target elevation values ​​corresponding to each spatial grid.

[0120] In this implementation, after completing detail enhancement and smoothing encryption processing on closed polygons to obtain polygonal surfaces, a final digital elevation model can be constructed based on the polygonal surfaces.

[0121] Optionally, a spatial grid (or grid) is a basic structure for constructing a digital elevation model, and the spatial grid divides the three-dimensional space into a series of regular two-dimensional units.

[0122] For example, a spatial grid is first created based on the filled polygonal surface. The size and shape of this spatial grid depends on the resolution and detail requirements of the desired digital elevation model. Typically, the spatial grid is designed to be square or rectangular, large enough to capture subtle changes in the terrain but not too dense to avoid unnecessary data redundancy. The purpose of creating a spatial grid is to provide a structured framework for storing and representing elevation information within the area to be driven.

[0123] Optionally, after the spatial grid is created, the elevation point information on the polygonal surface can be combined with the spatial grid, that is, the elevation value of each elevation point can be assigned to the spatial grid cell where the corresponding elevation point is located or the nearest spatial grid cell. The nearest neighbor matching method is used here to assign the elevation value of each elevation point to the spatial grid cell where the corresponding elevation point is located or the nearest spatial grid cell. In other words, each spatial grid cell will obtain the elevation value of the elevation point closest to it. This process ensures that the elevation values ​​on the spatial grid can accurately reflect the terrain characteristics of the polygonal surface, and through the previous detail enhancement processing, these elevation values ​​have already included the slight changes and noise of the terrain, thereby improving the realism and complexity of the digital elevation model.

[0124] Optionally, after assigning elevation values ​​to each spatial grid, in order to further improve the accuracy and reliability of the digital elevation model, the elevation values ​​of these spatial grids need to be locally optimized and smoothed to obtain the target elevation value corresponding to the spatial grid. This step can be completed using various mathematical methods, such as local mean or weighted mean calculation, to ensure that the elevation values ​​between adjacent grid cells transition smoothly without sudden changes or outliers.

[0125] Optionally, after obtaining the target elevation value corresponding to each spatial grid, a digital elevation model can be constructed based on the target elevation value corresponding to each spatial grid unit. For example, each spatial grid unit together with its elevation information is organized into an overall data structure, generally represented in the form of a grid or contour line. Among them, the construction of the digital elevation model also needs to consider the data format and storage method to ensure effective management and efficient access to the data. After the construction is completed, the digital elevation model will fully reflect the terrain elevation of the area to be driven, including various terrain features such as flat land, hills, and hillsides, providing detailed terrain information support for the path planning, obstacle avoidance, and dynamic simulation of unmanned vehicles.

[0126] In this step, a digital elevation model that meets the requirements of the mining environment can be efficiently constructed from the filled polygonal surfaces, significantly improving the density, accuracy, and reliability of the terrain data. Such a digital elevation model not only accurately reproduces terrain features but also allows the level of detail to be flexibly adjusted according to the actual scenario, greatly enhancing the practicality and accuracy of the unmanned driving simulation system. In addition, the spatial grid-based construction method is easily compatible with existing Geographic Information System (GIS) software, facilitating data analysis and visualization.

[0127] As an optional implementation, the elevation values ​​corresponding to the elevation points on the polygonal surface are assigned to the corresponding spatial grids to obtain the target elevation values ​​corresponding to each spatial grid, including: determining the elevation points corresponding to each spatial grid on the polygonal surface; determining the distance weights of the elevation points corresponding to each spatial grid based on the distances between the elevation points corresponding to each spatial grid and the center points of the corresponding spatial grids; and performing weighted average calculation of the elevation values ​​of the elevation points corresponding to each spatial grid based on the distance weights of the elevation points corresponding to each spatial grid to obtain the target elevation values ​​corresponding to each spatial grid.

[0128] In this embodiment, when assigning the elevation values ​​corresponding to the elevation points on the polygonal surface to the corresponding spatial grids, the elevation points corresponding to each spatial grid on the polygonal surface can be determined, and the distance between the elevation points in each spatial grid and the center point of the corresponding spatial grid can be determined, and then the distance weights are assigned to the elevation points corresponding to the spatial grids according to the distances, wherein the closer the elevation point is to the center point of the spatial grid, the higher the corresponding distance weight, and conversely, the farther the elevation point is from the center line of the spatial grid, the smaller the corresponding distance weight.

[0129] Optionally, after determining the distance weights for the elevation points corresponding to each spatial grid, a weighted average of the elevation values ​​of each elevation point within the empty grid is calculated to determine the target elevation value for each spatial grid. This weighted average process can make the elevation value of the grid point closer to the actual terrain height at the center of the grid, while also taking into account the influence of surrounding elevation points, thereby improving the accuracy and authenticity of the digital elevation model.

[0130] Alternatively, using the above method, each spatial grid will have a target elevation value that integrates the information of all elevation points within the grid area, thus completing the transformation from irregularly distributed elevation points to elevation values ​​on a regular grid. Each spatial grid can then be used to construct a digital elevation model of the area to be driven.

[0131] In this step, the target elevation value of each spatial grid is determined by performing a weighted average calculation on the elevation values ​​in each spatial grid. This not only retains the original information of the elevation points on the polygonal surface, but also uses the weight distribution mechanism to reasonably distribute it to the entire spatial grid, effectively reducing the information loss in the process of building the digital elevation model and improving the continuity and smoothness of the data.

[0132] As an optional implementation, the method further includes: establishing a spatial index for each spatial grid on the polygonal surface, wherein the spatial index is used to query the target elevation value corresponding to the spatial grid.

[0133] In this embodiment, a spatial index is a data structure used to accelerate the search and access of geospatial data. Spatial indexes are established for each spatial grid on a polygonal surface to quickly locate and query the elevation values ​​of each spatial grid. This is because the construction of a digital elevation model involves a large amount of spatial data processing. Without an indexing mechanism, directly traversing and querying the elevation values ​​of each spatial grid would be extremely time-consuming, significantly reducing overall processing speed and efficiency.

[0134] Optionally, when establishing a spatial index, you can choose an appropriate spatial index type based on the distribution of the spatial grid on the polygonal surface. For example, when using a quadtree index, the polygonal surface is divided into four sub-regions, each of which is further divided into four smaller sub-regions, and so on recursively until the desired level of accuracy is achieved. Each node stores the spatial grid information of the area it covers, including elevation values ​​and other relevant attributes.

[0135] Optionally, for each layer of index nodes, the elevation information of all spatial grids covered by it is stored for easy subsequent query. This information may include the coordinate range, average elevation value, minimum elevation value, maximum elevation value, etc. of the grid, which is not specifically limited here.

[0136] Optionally, once the spatial index is established, the elevation values ​​of any spatial grid can be quickly queried through the index structure. This query is usually much faster than searching directly in a large unindexed dataset.

[0137] This step, by establishing a spatial index, addresses the efficiency bottleneck associated with processing large amounts of spatial data. This is particularly true in mining environments, where digital elevation models can contain tens of thousands or even more spatial grids. Spatial indexing enables near-instant querying and accessing elevation values, which is crucial for real-time simulation and data analysis, significantly improving the performance and responsiveness of the unmanned driving simulation system.

[0138] The technical solutions of the embodiments of the present invention are described below with reference to preferred implementation methods.

[0139] At present, in the field of unmanned driving in mines, the common methods for constructing digital elevation models are mainly the following:

[0140] (1) High-density point cloud data of the ground surface of the mining scene is obtained through radar, and then a point cloud processing algorithm is used to construct a highly refined digital elevation model. However, the implementation cost of this method is high. It not only requires expensive hardware equipment (such as lidar), but also has a high dependence on the real environment. It is impossible to quickly generate the required model without real environment data.

[0141] (2) Professionals manually draw terrain elevation information based on topographic maps or field survey results, and then convert it into a digital elevation model. Although this method can ensure the customization and accuracy of the model, it is completely dependent on manual operation, has low construction efficiency, and cannot be automated, making it difficult to meet the needs of large-scale simulation or real-time updates.

[0142] (3) Using artificial intelligence and machine learning technology to automatically generate digital elevation models through training models. This method has obvious advantages in generation speed and degree of automation, but it requires a large amount of training data, and the generated terrain description may not be accurate enough, especially in areas with high terrain complexity, such as mining environments, where the details and authenticity of the model are difficult to effectively guarantee.

[0143] When the above-mentioned digital elevation model generation method is applied in a mining environment, it usually exposes technical problems such as high cost, low efficiency, and difficulty in automation, which limits the simulation and development process of mine unmanned driving technology.

[0144] However, the embodiment of the present application proposes a method for generating a digital elevation model. First, the vector map information of the area to be driven by the vehicle is read and parsed, and the vector map of the area to be driven is divided according to the area type; the borders of the divided polygons are smoothed in elevation, and the elevation points of the borders are encrypted using linear interpolation technology to reduce holes; the polygonal faces of the polygons are filled with scan lines; and a digital elevation model of the area to be driven is constructed based on the filled polygonal faces. That is to say, in the embodiment of the present application, the digital elevation model is consistent with the vector map information, which reduces the uncertainty in the simulation environment; and the polygons corresponding to the area to be driven are adjusted according to actual needs, so that a digital elevation model that meets the use needs can be constructed, thereby improving construction efficiency.

[0145] Figure 2 is a flow chart of another method for generating a digital elevation model according to an embodiment of the present invention, such as Figure 2 As shown, the method for generating the digital elevation model includes the following steps:

[0146] Step S201: Acquire vector map data of the mining area.

[0147] In this embodiment, vector map data of the mining area where the vehicle is to travel is read, and the coordinates of each elevation point and the corresponding terrain feature information in the mining area are parsed from the vector map data. Subsequently, based on the terrain feature information corresponding to each elevation point, the vector map corresponding to the mining area is segmented into multiple polygons, where the multiple polygons are closed and have no self-intersections.

[0148] Step S202: pre-process the vector map data.

[0149] In this embodiment, after the vector map corresponding to the mining area is divided into a plurality of polygons, the borders of the respective polygons may be smoothed according to the polygon segmentation results.

[0150] Alternatively, since the elevation points on the polygonal borders are key to connecting adjacent terrain areas during the digital elevation model generation process, if the elevation data at these points contains anomalies, such as sudden elevation jumps or discontinuities, these anomalies will result in unnatural terrain features in the final DEM, such as sharp bumps or depressions. This is inconsistent with the actual terrain and may affect subsequent terrain analysis, path planning, and simulation quality. Therefore, the polygonal borders need to be smoothed.

[0151] For example, within any polygon, distance weights are used to smooth the polygon's border elevations to prevent anomalous elevation points from causing abnormal bumps in the result. First, the set of neighboring elevation points for each border point is determined, typically within a predefined neighborhood. Next, the Euclidean distance from each border point to its neighboring elevation points is calculated. The choice of neighborhood range depends on the terrain complexity and the desired DEM accuracy to ensure that each point has sufficient neighboring elevation points to participate in the smoothing process. Based on the calculated distances, each neighboring elevation point is assigned a weight. Weights are typically calculated using an inverse relationship: closer points receive greater weights, while farther points receive smaller weights. The distance weights are then used to weight the average of the neighboring elevation points to obtain the adjusted elevation value for the border point. This process is repeated for each point on the polygon's border until the desired smoothing effect is achieved or a specified error bound is met. The number of iterations and error bounds depend on the terrain complexity and the desired DEM accuracy to ensure that all border points are appropriately adjusted to avoid the influence of outliers.

[0152] Optionally, during the elevation smoothing process, statistical methods (such as median filtering and standard deviation analysis) can be combined to detect outliers. Once the elevation value of a point is found to be significantly deviated from the elevation values ​​of its neighboring elevation points, it can be considered an outlier and corrected using the elevation values ​​of the neighboring elevation points to further ensure the continuity and rationality of the polygon border elevation.

[0153] Optionally, after performing elevation smoothing on multiple polygons, the borders of the polygons may be further encrypted by linear interpolation to reduce holes on the edges of the polygons.

[0154] For example, a polygon border is composed of a series of discrete points representing the boundaries of the terrain. Border densification involves inserting additional points between existing points to increase the point density on the border. This allows for a more detailed vector map depiction, resulting in a smoother and more coherent terrain surface during the subsequent digital elevation model generation process. Linear interpolation is used to estimate unknown values ​​between two known points. During polygon border densification, if the elevation between two adjacent points is known, linear interpolation can be used to estimate the elevation of any location between them. For example, if the straight line between two points represents a path of elevation change, the elevation of the new point inserted between them can be proportionally distributed to the elevation of the two points based on their position. If the point density on the polygon border is insufficient, using scanlines to fill the polygon surface may leave holes or discontinuous areas within the polygon. These holes are particularly noticeable when the terrain changes significantly. Densifying the border, i.e., increasing the point density on the border, provides more points for reference when scanning the polygon surface, thus reducing the occurrence of internal holes and making the resulting polygon surface more complete and coherent.

[0155] This step effectively processes the elevation data on the polygonal border, eliminating or mitigating model defects caused by unusual elevation points. The polygonal border is then encrypted, resulting in a smoother, more realistic, and more user-friendly digital elevation model. This method not only improves the efficiency of building digital elevation models but also ensures accurate representation of terrain features, making it particularly suitable for environments with complex terrain, such as mines.

[0156] Step S203: determining polygonal surfaces based on the pre-processed vector map data.

[0157] In this embodiment, after performing elevation smoothing and encryption processing on the polygonal border, a scan line algorithm may be further used to fill the interior of the polygon to obtain a polygonal face.

[0158] For example, when filling the interior of a polygon using a scanline algorithm, the entire polygon area is typically scanned line by line along a single direction (e.g., perpendicular to the polygon's edges, such as from left to right or bottom to top). In a scanline algorithm, the elevation of the polygon face is determined by identifying the intersection of the scanline with the polygon's bounding box. When the scanline intersects the bounding box, the elevation of that intersection is calculated.

[0159] Optionally, for each fill point between two intersection points, its elevation value is usually calculated by linear interpolation based on the elevation values ​​of the two intersection points. For example, if the scan line moves from a lower elevation intersection to a higher mid-range intersection, the elevation value of the fill point will change linearly from low to high, and vice versa. This process ensures the continuity and natural transition of the terrain surface. For example, suppose the scan line intersects with the encrypted border at x=25, the elevation of the intersection is 125, and then intersects again at x=75, the elevation of the intersection is 175. Then, the elevation value of each point between x=25 and x=75 (i.e., the fill point) will be calculated by linear interpolation based on its specific position, thus forming a terrain surface that changes continuously from 125 to 175.

[0160] Optionally, after the polygonal border is scanned according to the scan line algorithm, the polygonal border may be filled to obtain a polygonal face.

[0161] Step S204: constructing a digital elevation model based on the polygonal surface.

[0162] In this embodiment, after the polygonal surface is obtained according to step S203, a digital elevation model can be constructed according to the polygonal surface information.

[0163] Optionally, a spatial grid is constructed based on the polygonal surface, and then a spatial index is established for each spatial grid. The elevation value of each spatial grid is calculated using the nearest neighbor matching method, and each spatial grid is elevation smoothed to obtain a digital elevation model corresponding to the mining area.

[0164] In steps S201 through S204, the DEM is directly generated using vector map information, ensuring that terrain features are consistent with the elevation data marked on the map. This significantly reduces simulation system uncertainty caused by mismatches between map and terrain information. In a mining environment, this consistency is crucial for the correct simulation of vehicle dynamics models, ensuring the reliability and effectiveness of simulation results.

[0165] According to an embodiment of the present invention, a device for generating a digital elevation model is also provided. It should be noted that the device for generating a digital elevation model can be used to execute the method for generating a digital elevation model in an embodiment of the present invention.

[0166] Figure 3 3 is a schematic diagram of a digital elevation model generation device according to an embodiment of the present invention. The digital elevation model generation device 300 includes: an acquisition unit 301, a determination unit 302, a division unit 303, a processing unit 304, a scanning unit 305, and a construction unit 306.

[0167] The acquisition unit 301 is used to acquire vector map information of the area where the vehicle is to travel, wherein the vector map information is used to represent the terrain features of the area to be traveled, and the area to be traveled includes multiple elevation points.

[0168] The determining unit 302 is configured to determine the terrain feature types corresponding to the plurality of elevation points based on the vector map information.

[0169] The dividing unit 303 is configured to divide the vector map corresponding to the area to be driven into at least one closed polygon based on the terrain feature types corresponding to the plurality of elevation points, wherein the terrain feature types corresponding to the elevation points in a closed polygon are consistent.

[0170] The processing unit 304 is configured to perform elevation smoothing on the border of the at least one closed polygon to obtain at least one smoothed closed polygon.

[0171] The scanning unit 305 is configured to scan the at least one smoothed closed polygon and fill the polygonal surface of the at least one closed polygon with scan lines.

[0172] The construction unit 306 is configured to construct a digital elevation model corresponding to the area to be traveled based on the filled polygonal surface.

[0173] Optionally, the determination unit 302 also includes: a parsing module, used to parse the vector map information and determine the attribute information corresponding to multiple elevation points respectively, wherein the attribute information is used to at least represent the location information and elevation information of the corresponding elevation points; a first determination module, used to determine the terrain feature types corresponding to the multiple elevation points based on the attribute information corresponding to the multiple elevation points.

[0174] Optionally, the division unit 303 also includes: an identification module, which is used to identify elevation points with the same terrain feature type from multiple elevation points based on the terrain feature types corresponding to the multiple elevation points; a division module, which is used to divide the elevation points with the same terrain feature type into at least one area based on the position information of the elevation points with the same terrain feature type; and a generation module, which is used to generate at least one closed polygon based on the boundary of at least one area.

[0175] Optionally, the processing unit 304 also includes: a second determination module for determining the distance between each elevation point on the border of at least one closed polygon and each adjacent elevation point within a preset range; a first allocation module for allocating distance weights to each adjacent elevation point based on the distance between each elevation point and each adjacent elevation point within the preset range, wherein the distance weight corresponding to the adjacent elevation point that is closer to each elevation point is greater; an adjustment module for adjusting the elevation value corresponding to each elevation point based on the elevation values ​​of each adjacent elevation point corresponding to each elevation point and the distance weight corresponding to each adjacent elevation point to obtain a target elevation value corresponding to each elevation point; a processing module for performing elevation smoothing processing on the border of at least one closed polygon based on the target elevation value corresponding to each elevation point to obtain at least one smoothed closed polygon.

[0176] Optionally, the adjustment module is also used to: perform weighted averaging on the elevation values ​​of each adjacent elevation point corresponding to each elevation point and the distance weights corresponding to each adjacent elevation point to obtain a weighted average value corresponding to each elevation point; and use the weighted average value corresponding to each elevation point to replace the elevation value corresponding to each elevation point to obtain a target elevation value corresponding to each elevation point.

[0177] Optionally, the scanning unit 305 also includes: an enhancement module for performing detail enhancement processing on at least one smoothed closed polygon to obtain at least one enhanced closed polygon; an encryption module for encrypting the border of at least one enhanced closed polygon to obtain at least one encrypted closed polygon; and a scanning module for scanning the at least one encrypted closed polygon to fill the polygonal surface of the at least one closed polygon.

[0178] Optionally, the enhancement module is also used to: determine the noise type corresponding to at least one closed polygon based on the terrain feature types corresponding to multiple elevation points in at least one smoothed closed polygon; generate noise values ​​matching the multiple elevation points based on the noise type and the position information corresponding to the multiple elevation points; add the noise value to the elevation value of the corresponding elevation point, enhance the terrain features of the multiple elevation points, and obtain at least one enhanced closed polygon.

[0179] Optionally, the encryption module is also used to: determine the encryption interval for encrypting the border of at least one enhanced closed polygon; insert at least one new elevation point on each border of at least one enhanced closed polygon based on the encryption interval through linear interpolation technology; determine the elevation value corresponding to the inserted at least one new elevation point; add the at least one new elevation point and the elevation value corresponding to the at least one new elevation point to the border of at least one closed polygon to obtain the encrypted at least one closed polygon.

[0180] Optionally, the scanning module is further configured to: scan the at least one encrypted closed polygon using a scanning line to fill the polygonal surface of the at least one closed polygon.

[0181] Optionally, the scanning module is also used to perform the following steps to scan at least one encrypted closed polygon using a scanning line to fill the polygonal surface of at least one closed polygon: a scanning step of controlling the scanning line to scan at least one encrypted closed polygon in one direction; a determining step of determining the elevation value corresponding to the intersection in response to the scanning line intersecting with the border of at least one encrypted closed polygon; an adding step of adding the intersection and the elevation value corresponding to the intersection to the at least one encrypted closed polygon; and repeating the scanning step, the determining step and the adding step until the polygonal surface of the at least one encrypted closed polygon is completely filled.

[0182] Optionally, the construction unit 306 includes: a first construction module, used to construct a spatial grid on the filled polygonal surface; a second allocation module, used to assign the elevation values ​​corresponding to the elevation points on the polygonal surface to the corresponding spatial grids to obtain the target elevation values ​​corresponding to each spatial grid; and a second construction module, used to construct a digital elevation model corresponding to the area to be driven based on the target elevation values ​​corresponding to each spatial grid.

[0183] Optionally, the second allocation module is also used to: determine the elevation points corresponding to each spatial grid on the polygonal surface; determine the distance weight of the elevation points corresponding to each spatial grid based on the distance between the elevation points corresponding to each spatial grid and the center point of the corresponding spatial grid; based on the distance weight of the elevation points corresponding to each spatial grid, perform weighted average calculation on the elevation values ​​of the elevation points corresponding to each spatial grid to obtain the target elevation value corresponding to each spatial grid.

[0184] Optionally, the apparatus 300 for generating a digital elevation model is further configured to: establish a spatial index for each spatial grid on the polygonal surface, wherein the spatial index is used to query a target elevation value corresponding to the spatial grid.

[0185] In this device, vector map information of the area to be traveled is directly used to determine the terrain feature type corresponding to each elevation point in the area to be traveled. Based on the terrain feature type corresponding to each elevation point in the area to be traveled, the vector map is accurately divided into a series of closed polygons containing elevation points of the same terrain feature. This process does not require additional hardware equipment to collect data, significantly reducing initial costs. Next, by performing elevation smoothing on the borders of the closed polygons, abnormal changes in the elevation data are eliminated, thereby improving the visual and physical consistency of the subsequently constructed digital elevation model. Subsequently, efficient scan line technology is used to fill the polygon faces to quickly construct a complete terrain surface. Furthermore, the terrain features of the digital elevation model generated directly from the vector map information are strictly consistent with the original map data, while retaining rich terrain details and avoiding the uncertainty introduced by information mismatches. This greatly enhances the reliability and authenticity of the simulation system, thereby improving the accuracy and efficiency of digital elevation model construction, thereby solving the technical problem of low digital elevation model generation efficiency.

[0186] An embodiment of the present application further provides an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods of various embodiments of the present invention when running.

[0187] An embodiment of the present application further provides a computer-readable storage medium, which includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0188] An embodiment of the present application further provides a computer program product, including a computer program, which implements the methods in various embodiments of the present invention when executed by a processor.

[0189] An embodiment of the present application further provides a computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0190] The embodiments of the present application further provide a computer program, which implements the methods in the above-mentioned embodiments of the present invention when executed by a processor.

[0191] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0192] In the above embodiments of the present invention, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0193] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. At another high level, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0194] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0195] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0196] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, etc. Various media that can store program codes.

[0197] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for generating a digital elevation model, characterized in that: Applied to a mining scenario, the method includes: Acquiring vector map information of an area to be traveled by the vehicle, wherein the vector map information is used to represent the terrain features of the area to be traveled, and the area to be traveled includes a plurality of elevation points; Based on the vector map information, determining the terrain feature types corresponding to the plurality of elevation points respectively; Based on the types of terrain features corresponding to the plurality of elevation points, the vector map corresponding to the area to be driven is divided into at least one closed polygon, wherein the types of terrain features corresponding to the elevation points in one closed polygon are consistent; Performing elevation smoothing on the border of the at least one closed polygon to obtain the at least one smoothed closed polygon; Scanning the at least one smoothed closed polygon, and filling the polygonal surface of the at least one closed polygon with scan lines; Based on the filled polygonal surface, a digital elevation model corresponding to the area to be traveled is constructed.

2. The method according to claim 1, characterized in that Determining, based on the vector map information, terrain feature types corresponding to the plurality of elevation points, respectively, includes: Parsing the vector map information to respectively determine attribute information corresponding to the plurality of elevation points, wherein the attribute information is used to at least represent position information and elevation information of the corresponding elevation points; Based on the attribute information corresponding to the plurality of elevation points, the terrain feature types corresponding to the plurality of elevation points are respectively determined.

3. The method according to claim 1, characterized in that Dividing the vector map corresponding to the area to be driven into at least one closed polygon based on the terrain feature types corresponding to the plurality of elevation points includes: Based on the terrain feature types corresponding to the plurality of elevation points, identifying elevation points with the same terrain feature type from the plurality of elevation points; Dividing the elevation points of the same terrain feature type into at least one area based on the position information of the elevation points of the same terrain feature type; At least one closed polygon is generated based on the boundary of the at least one region.

4. The method according to claim 1, wherein Performing elevation smoothing on the border of the at least one closed polygon to obtain at least one smoothed closed polygon includes: Determining the distance between each elevation point on the border of the at least one closed polygon and each adjacent elevation point within a preset range; Based on the distance between each elevation point and each adjacent elevation point within a preset range, a distance weight is assigned to each adjacent elevation point, wherein the distance weight corresponding to the adjacent elevation point that is closer to each elevation point is greater; Based on the elevation values ​​of each adjacent elevation point corresponding to each elevation point and the distance weights corresponding to each adjacent elevation point, the elevation value corresponding to each elevation point is adjusted to obtain a target elevation value corresponding to each elevation point; Based on the target elevation value corresponding to each elevation point, elevation smoothing is performed on the border of the at least one closed polygon to obtain at least one smoothed closed polygon.

5. The method according to claim 4, characterized in that Based on the elevation values ​​of each adjacent elevation point corresponding to each elevation point and the distance weights corresponding to each adjacent elevation point, the elevation value corresponding to each elevation point is adjusted to obtain the target elevation value corresponding to each elevation point, including: Performing weighted averaging on the elevation values ​​of each adjacent elevation point corresponding to each elevation point and the distance weights corresponding to each adjacent elevation point to obtain a weighted average value corresponding to each elevation point; The weighted average value corresponding to each elevation point is used to replace the elevation value corresponding to each elevation point, so as to obtain the target elevation value corresponding to each elevation point.

6. The method according to claim 1, characterized in that Scanning the at least one smoothed closed polygon to fill the polygonal surface of the at least one closed polygon includes: performing detail enhancement processing on the at least one smoothed closed polygon to obtain the at least one enhanced closed polygon; Encrypting the enhanced border of the at least one closed polygon to obtain the encrypted at least one closed polygon; The encrypted at least one closed polygon is scanned to fill the polygonal surface of the at least one closed polygon.

7. The method according to claim 6, characterized in that Performing detail enhancement processing on a plurality of elevation points in the at least one smoothed closed polygon to obtain the at least one enhanced closed polygon, comprising: Determining a noise type corresponding to the at least one closed polygon based on terrain feature types corresponding to a plurality of elevation points in the at least one smoothed closed polygon; generating noise values ​​matching the plurality of elevation points based on the noise type and the position information corresponding to the plurality of elevation points; The noise value is added to the elevation value of the corresponding elevation point, and the terrain features of the plurality of elevation points are enhanced to obtain the at least one enhanced closed polygon.

8. The method according to claim 6, characterized in that Encrypting the enhanced border of the at least one closed polygon to obtain the encrypted at least one closed polygon includes: Determining an encryption interval for encrypting the enhanced border of the at least one closed polygon; Inserting at least one new elevation point on each border of the at least one closed polygon after enhancement based on the refined interval by a linear interpolation technique; Determining an elevation value corresponding to the inserted at least one new elevation point; The at least one new elevation point and the elevation value corresponding to the at least one new elevation point are added to the border of the at least one closed polygon to obtain the encrypted at least one closed polygon.

9. The method according to claim 6, characterized in that Scanning the encrypted at least one closed polygon to fill the polygonal surface of the at least one closed polygon includes: The encrypted at least one closed polygon is scanned using a scan line to fill the polygonal surface of the at least one closed polygon.

10. The method according to claim 9, characterized in that Scanning the encrypted at least one closed polygon using a scan line algorithm to fill the polygonal surface of the at least one closed polygon includes: A scanning step, controlling a scanning line to scan the at least one encrypted closed polygon in one direction; a determining step, in response to the scanning line intersecting with the encrypted border of the at least one closed polygon, determining an elevation value corresponding to the intersection point; An adding step of adding the intersection point and the elevation value corresponding to the intersection point to the at least one encrypted closed polygon; The scanning step, the determining step, and the adding step are repeatedly performed until the polygonal surface of the at least one closed polygon after encryption is completely filled.

11. The method according to claim 1, wherein Constructing a digital elevation model corresponding to the area to be driven based on the filled polygonal surface, including: Constructing a spatial grid on the filled polygonal surface; Assigning the elevation values ​​corresponding to the elevation points on the polygonal surface to the corresponding spatial grids to obtain the target elevation values ​​corresponding to the respective spatial grids; Based on the target elevation values ​​corresponding to the respective spatial grids, a digital elevation model corresponding to the area to be traveled is constructed.

12. The method according to claim 11, characterized in that Assigning the elevation values ​​corresponding to the elevation points on the polygonal surface to the corresponding spatial grids to obtain the target elevation values ​​corresponding to the spatial grids, including: Determining the elevation points corresponding to each spatial grid on the polygonal surface; Determining distance weights of the elevation points corresponding to the respective spatial grids based on distances between the elevation points corresponding to the respective spatial grids and the center points of the corresponding spatial grids; Based on the distance weights of the elevation points corresponding to the respective spatial grids, weighted average calculation is performed on the elevation values ​​of the elevation points corresponding to the respective spatial grids to obtain target elevation values ​​corresponding to the respective spatial grids.

13. The method according to claim 11, characterized in that The method further comprises: A spatial index is established for each spatial grid on the polygonal surface, wherein the spatial index is used to query the target elevation value corresponding to the spatial grid.

14. A device for generating a digital elevation model, characterized in that: Applied to mining scenarios, the device includes: An acquisition unit, configured to acquire vector map information of an area to be traveled by the vehicle, wherein the vector map information is used to represent the terrain features of the area to be traveled, and the area to be traveled includes a plurality of elevation points; a determining unit, configured to determine, based on the vector map information, the types of terrain features corresponding to the plurality of elevation points; a dividing unit, configured to divide the vector map corresponding to the area to be traveled into at least one closed polygon based on the types of terrain features corresponding to the plurality of elevation points, wherein the types of terrain features corresponding to the elevation points in a closed polygon are consistent; a processing unit, configured to perform elevation smoothing on the border of the at least one closed polygon to obtain the at least one smoothed closed polygon; a scanning unit, configured to scan the at least one smoothed closed polygon and fill the polygonal surface of the at least one closed polygon with a scanning line; The construction unit is used to construct a digital elevation model corresponding to the area to be traveled based on the filled polygonal surface.

Images