Road element elevation determination and high-definition map production method and device, medium

Abstract

The embodiment of the present disclosure discloses a road element elevation determination and high-definition map production method, device and medium, which comprises the following steps: determining a point cloud grid corresponding to a road region according to two-dimensional position information of the point cloud in the road region, wherein the point cloud grid is a grid of a predetermined shape divided according to two-dimensional positions; determining the elevation of the point cloud grid according to the elevation information of the point cloud in the point cloud grid; and calculating the elevation of a shape point of a road element according to the elevations of the point cloud grid where the shape point of the road element is located and the point cloud grid adjacent to the point cloud grid. The technical solution can make the calculated shape point of the road element more accurate, thereby making the produced three-dimensional road element more consistent with the point cloud, reducing manual work, and improving production efficiency.

Description

Technical Field

[0001] This disclosure relates to the field of map production technology, specifically to a method, apparatus, and medium for determining the elevation of road elements and producing high-precision maps. Background Technology

[0002] In the process of creating high-precision maps, the shapes of road features are first produced in a 2D production workshop. Then, in a 3D production workshop, elevation calculation services are used to calculate the elevation information of each road feature. If the elevation information is inaccurate, manual adjustments are required to align the elevation of the produced road features with their point cloud data. However, due to the complexity of the 3D production workshop environment, manual operations are inefficient. Therefore, improving the accuracy of elevation calculations to reduce the difficulty of manual operations has become a pressing technical problem. Summary of the Invention

[0003] To address the problems in related technologies, this disclosure provides a method, apparatus, and medium for determining the elevation of road elements and creating high-precision maps.

[0004] Firstly, this disclosure provides a method for determining the elevation of road elements.

[0005] Specifically, the method for determining the elevation of road elements includes:

[0006] Based on the two-dimensional position information of the point cloud within the road area, the point cloud grid corresponding to the road area is determined, wherein the point cloud grid is a grid of a predetermined shape divided according to the two-dimensional position.

[0007] The elevation of the point cloud grid is determined based on the elevation information of the point cloud within the grid.

[0008] The elevation of the shape point of the road element is calculated based on the elevation of the point cloud grid where the shape point of the road element is located and the elevation of the point cloud grid adjacent to it within the road area.

[0009] Secondly, this disclosure provides a method for creating high-precision maps, including:

[0010] Based on the original road data corresponding to the road area, generate two-dimensional road elements within the road area;

[0011] The elevation of the shape points of the two-dimensional road element is calculated using the method described in any one of the first aspects;

[0012] Using the elevations of two-dimensional road features and their shape points within the road area, three-dimensional road features within the road area are generated.

[0013] Thirdly, this disclosure provides a road element elevation determination device, including:

[0014] The grid determination module is configured to determine the point cloud grid corresponding to the road area based on the two-dimensional position information of the point cloud within the road area, wherein the point cloud grid is a grid of a predetermined shape divided according to the two-dimensional position.

[0015] The grid elevation determination module is configured to determine the elevation of the point cloud grid based on the elevation information of the point cloud within the grid.

[0016] The shape point elevation determination module is configured to calculate the elevation of the shape point of the road element based on the elevation of the point cloud grid where the shape point of the road element is located and the elevation of the point cloud grid adjacent to it within the road area.

[0017] Fourthly, this disclosure provides a high-precision map production apparatus, comprising:

[0018] The two-dimensional production module is configured to produce two-dimensional road elements within the road area based on the original road data corresponding to the road area.

[0019] The elevation calculation module is configured to calculate the elevation of the shape points of the two-dimensional road element using the method described in any one of the first aspects;

[0020] The 3D production module is configured to produce 3D road features within the road area using the elevations of 2D road features and their shape points within the road area.

[0021] Fifthly, embodiments of this disclosure provide an electronic device including a memory and a processor, wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method as described in any one of the first or second aspects.

[0022] In a sixth aspect, embodiments of this disclosure provide a computer-readable storage medium having computer instructions stored thereon that, when executed by a processor, implement the method as described in any one of the first or second aspects.

[0023] In a seventh aspect, this disclosure provides a computer program product including computer instructions that, when executed by a processor, implement the method steps as described in any one of the first or second aspects.

[0024] According to the technical solution provided in this disclosure, the point cloud grid corresponding to the road area can be determined based on the point cloud data within the road area; and the elevation of the point cloud grid can be determined based on the elevation information of the point cloud within the point cloud grid. Then, the elevation of the shape point of the road element can be calculated based on the elevation of the point cloud grid where the shape point of the road element is located and the elevation of the point cloud grid adjacent to it. The shape point of the road element calculated in this way is more accurate, making the produced three-dimensional road element fit its point cloud better, reducing manual work, and thus improving production efficiency.

[0025] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0026] Other features, objects, and advantages of this disclosure will become more apparent from the following detailed description of non-limiting embodiments, taken in conjunction with the accompanying drawings. In the drawings:

[0027] Figure 1 A flowchart illustrating a method for determining the elevation of road elements according to an embodiment of the present disclosure is shown;

[0028] Figure 2A A schematic diagram of a mesh is shown according to an embodiment of the present disclosure;

[0029] Figure 2B A schematic diagram of planar fitting according to an embodiment of the present disclosure is shown;

[0030] Figure 3A A flowchart illustrating a method for creating high-precision maps according to an embodiment of the present disclosure;

[0031] Figure 3B A schematic diagram illustrating an application scenario of a method for determining road element elevation and creating high-precision maps according to an embodiment of the present disclosure is shown.

[0032] Figure 4A A structural block diagram of a road element elevation determination device according to an embodiment of the present disclosure is shown;

[0033] Figure 4B A structural block diagram of a high-precision map-making apparatus according to an embodiment of the present disclosure is shown;

[0034] Figure 5 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown;

[0035] Figure 6 A schematic diagram of the structure of a computer system suitable for implementing the method according to embodiments of the present disclosure is shown. Detailed Implementation

[0036] In the following, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings to enable those skilled in the art to readily implement them. Furthermore, for clarity, portions unrelated to the description of exemplary embodiments have been omitted from the drawings.

[0037] In this disclosure, it should be understood that terms such as “comprising” or “having” are intended to indicate the presence of features, figures, steps, behaviors, components, parts or combinations thereof disclosed in this specification, and are not intended to exclude the possibility of the presence or addition of one or more other features, figures, steps, behaviors, components, parts or combinations thereof.

[0038] It should also be noted that, unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other. This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0039] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0040] As mentioned above, the production process of high-precision maps requires first creating the shapes of road features in a 2D production workshop. Then, in a 3D production workshop, elevation calculation services are used to calculate the elevation information of each road feature. If the elevation information is inaccurate, manual adjustments are needed to align the elevation of the produced road features with their point cloud. However, due to the complexity of the 3D production workshop environment and the low efficiency of manual operations, improving the accuracy of elevation calculations to reduce the difficulty of manual work has become a pressing technical problem.

[0041] This disclosure provides a method for determining the elevation of road elements. This method can comprehensively calculate the elevation of the shape points of road elements by combining the elevation of the grid where the shape points of the road elements are located within the road area and the elevation of the grids adjacent to them. The elevation of the grid is obtained by comprehensively calculating the elevation of the point cloud within the grid. The shape points of the road elements calculated in this way are more accurate, making the produced three-dimensional road elements fit their point clouds better, reducing manual work, and thus improving production efficiency.

[0042] Figure 1 A flowchart illustrating a method for determining the elevation of road features according to an embodiment of this disclosure is shown. Figure 1 As shown, the method for determining the elevation of road elements includes the following steps S101-S103:

[0043] In step S101, the point cloud grid corresponding to the road area is determined based on the two-dimensional position information of the point cloud within the road area, and the point cloud grid is divided according to a predetermined grid division rule.

[0044] In step S102, the elevation of the point cloud grid is determined based on the elevation information of the point cloud within the point cloud grid.

[0045] In step S103, the elevation of the shape point of the road element is calculated based on the elevation of the point cloud grid where the shape point of the road element is located and the elevation of the point cloud grid adjacent to it within the road area.

[0046] In one possible implementation, the road feature elevation determination method is applicable to devices such as computers, computing devices, servers, and server clusters that can perform road feature elevation determination.

[0047] In one possible implementation, the point cloud within the road area is a collection of massive points on the surface of various road elements collected by point cloud acquisition devices such as lidar within the road area. This includes the point cloud of each road element, which includes the two-dimensional location information (such as latitude and longitude information) and elevation information of these massive points.

[0048] In one possible implementation, the point cloud mesh is a mesh of a predetermined shape divided according to two-dimensional location. For example, the Geohash algorithm treats the Earth as a two-dimensional plane and recursively decomposes the plane into a series of mesh levels. Geohash meshes are typically rectangular or square meshes with 12 levels, ranging from approximately 5000 km to approximately 3 cm. Google S2 uses square meshes, providing 30 levels, ranging from approximately 8000 km to approximately 1 cm, while Uber H3 uses hexagonal meshes, providing 15 levels, ranging from approximately 1107 km to approximately 0.5 m. The mesh size generated by the Geohash algorithm is moderate, while also possessing multi-level coverage capability. Furthermore, its encoding is relatively simple and universal, and it exhibits local order preservation, making it more suitable for the scenario of determining the elevation of road elements within a high-precision road surface in this implementation. Therefore, in this implementation, the Geohash mesh can be selected as the point cloud mesh.

[0049] In one possible implementation, the road network grid where the point cloud in the road area is located can be determined based on the two-dimensional location information of the point cloud in the road area. For example, GeoHash calculation can be performed on the two-dimensional location information of each point in the point cloud to obtain the GeoHash grid code corresponding to each point. The GeoHash grid corresponding to the grid code is the point cloud grid where the point is located. In this way, the point cloud grid corresponding to the point cloud in the road area can be obtained, that is, the point cloud grid corresponding to the road area.

[0050] In one possible implementation, for each point cloud grid, the elevation of the point cloud grid can be calculated by averaging the elevations of all points in the point cloud within the grid. It should be noted that, to ensure accuracy, outlier points in the point cloud grid can be filtered out first, and then the normal points in the point cloud grid can be averaged to obtain the elevation of the point cloud grid. Outlier points refer to convex points in the point cloud, including points with large elevation differences from other points and convex points at the edge of the point cloud.

[0051] In one possible implementation, the 2D production workshop has already created the two-dimensional shape of the road elements in the road area based on the original collected data such as trajectory data, image data, and point cloud data within the road area. In the 3D production workshop, it is necessary to calculate the elevation of the shape points of the road elements in the road area in order to produce the three-dimensional road elements. The shape points of the road elements refer to the points that can determine the shape of the road elements. They can be points on the two-dimensional shape curve of the road elements. The shape points in the road area include the shape points of linear elements such as lane lines, as well as the shape points of planar elements such as text and guide strips on the road.

[0052] In one possible implementation, the elevation of the point cloud grid containing the shape point of the road feature, as well as the elevation of the point cloud grid adjacent to the shape point, can be obtained. The elevation of the shape point is determined by combining the elevations of these point cloud grids. For example, the elevation of the shape point can be calculated as the average of the elevations of the point cloud grid containing the shape point of the road feature and the elevations of the point cloud grid adjacent to the shape point.

[0053] This embodiment can determine the point cloud grid corresponding to the road area based on the point cloud data within the road area; and determine the elevation of the point cloud grid based on the elevation information of the point cloud within the point cloud grid. Then, based on the elevation of the point cloud grid where the shape point of the road element is located and the elevation of the point cloud grid adjacent to it, the elevation of the shape point of the road element is calculated. The shape point of the road element calculated in this way is more accurate, making the produced 3D road element fit its point cloud better, reducing manual work, and thus improving production efficiency.

[0054] In one possible implementation, calculating the elevation of the shape point of the road feature based on the elevation of the point cloud grid where the shape point of the road feature is located and the elevation of the point cloud grids adjacent to it includes:

[0055] Recall the point cloud grid containing the shape points of the road element and the point cloud grid adjacent to it from the point cloud grid of the predetermined level corresponding to the road area;

[0056] In response to the fact that the number of recalled point cloud meshes is greater than 0, usable point cloud meshes are selected from the recalled point cloud meshes according to the predetermined filtering rules.

[0057] The elevation of the shape points of the road feature is determined based on the elevation of the available point cloud mesh.

[0058] In this implementation, if the point cloud grid is a geohash grid, considering the geographical area covered by geohash grids at each level, to ensure the accuracy of the elevation calculation of the shape point from the point cloud grid, the coverage area of ​​the point cloud grid containing the shape point and its adjacent point cloud grids should be appropriate. If it is too large, too many point clouds will be covered, and most of the point clouds will not be near the shape point, resulting in inaccurate elevation calculations. If it is too small, too few point clouds will be covered, and the calculated elevation will also be inaccurate. For example, the predetermined level can be a level 10 geohash grid or a level 11 geohash grid. Based on the two-dimensional location of the shape point of the road feature, the point cloud grid containing the shape point of the road feature and its adjacent point cloud grids can be retrieved from the ground grid of the predetermined level corresponding to the road area; for example, Figure 2A A schematic diagram of a mesh according to an embodiment of the present disclosure is shown, such as... Figure 2A The grid shown is a grid of a predetermined level. Point clouds are collected within the areas covered by grids 21, 22, 23, and 24. Based on the two-dimensional location information of these point clouds, the predetermined level point cloud grids corresponding to the road area can be determined as grids 21, 22, 23, and 24. Point cloud grid 23, where the road shape point A is located, can be recalled, as well as the point cloud grids adjacent to point cloud grid 23, namely grids 21, 22, and 24.

[0059] In this implementation, if no point cloud is collected in the area where the shape point of the road element is located and its surrounding area, the point cloud grid of the predetermined level corresponding to the road area may not contain the point cloud grid where the shape point of the road element is located and its adjacent point cloud grids. In this case, the number of recalled point cloud grids may be 0. If point cloud is collected in the area where the shape point of the road element is located or its surrounding area, the point cloud grid of the predetermined level corresponding to the road area will include the point cloud grid where the shape point of the road element is located and its adjacent point cloud grids. In this case, the number of recalled point cloud grids will not be 0, but greater than 0. For example, Figure 2A As shown, the number of grids recalled is 4.

[0060] In this embodiment, if the number of recalled point cloud grids is greater than 0, in order to ensure that the elevation of the calculated shape points is more accurate, it is necessary to improve the accuracy of the elevation of the point cloud grids used for calculation. Therefore, according to a predetermined filtering rule, point cloud grids with relatively accurate elevations can be selected from the recalled point cloud grids as usable point cloud grids, and then the elevation of the shape points of the road element can be calculated based on the elevation of these usable point cloud grids.

[0061] This embodiment can limit the level of the point cloud grid where the shape point of the recalled road element is located and the point cloud grid adjacent to it. This limits the coverage of the recalled point cloud grid, making the elevation of the recalled point cloud grid more accurately reflect the elevation of the shape point. At the same time, the recalled point cloud grid is filtered to obtain usable point cloud grids with accurate elevations. Then, the elevation of the shape point of the road element is determined based on the elevation of the usable point cloud grids. This can improve the accuracy of the elevation of the shape point of the road element.

[0062] In one possible implementation, recalling the point cloud grid containing the shape points of the road element and its adjacent point cloud grids from the point cloud grid of a predetermined level corresponding to the road area includes:

[0063] Recall the point cloud grid containing the shape points of the road element and the point cloud grid adjacent to them from the point cloud grid at the initial recall level corresponding to the road area;

[0064] In response to the number of recalled point cloud grids being 0, the point cloud grid containing the shape point of the road element and its adjacent point cloud grids are recalled from the point cloud grid of the previous level of the initial recall level, until the number of recalled point cloud grids is greater than 0 or the recall level reaches a preset level threshold, wherein the coverage area of ​​the previous level point cloud grid is larger than that of the point cloud grid of the initial recall level.

[0065] In this implementation, each grid level covers a different geographical area; the higher the grid level, the larger the geographical area it covers. Assuming the initial recall level is a geohash grid of level 10, the geographical area covered by the ground grid of this initial recall level is relatively small. The area covered by the point cloud grid of the initial recall level where the shape point of the road element is located, as well as the point cloud grid of the adjacent initial recall level, is also relatively small. Point clouds may not have been collected in this area, and the corresponding point cloud grid cannot be recalled from the point cloud grid of the initial recall level corresponding to the road area. In this case, the recall area can be expanded to recall the point cloud grid where the shape point of the road element is located, as well as the point cloud grid of the adjacent geohash grid of the next higher level, i.e., level 9. Of course, if the recall area is expanded and the point cloud grid is recalled (i.e., the number of recalled point cloud grids is greater than 0), the recall area will not be expanded further. If the point cloud grid is not recalled after the large recall area, it can be determined whether the current recall level has reached the preset level threshold. The preset level threshold is used to limit the recall range. This is because if the range is too large, the recalled grids will be far away from the shape point and cannot represent the actual elevation of the current position of the shape point, which does not meet the error requirements. Therefore, when the recall level reaches the preset level threshold, the recall area will not be expanded further even if the point cloud grid is not recalled.

[0066] In this implementation, when recalling point cloud meshes, the recall gradually expands from the initial recall level to the next higher level with a larger coverage area. If the number of point cloud meshes recalled at the initial recall level is greater than 0, the recall of meshes in a larger area will not continue, ensuring that the recalled point cloud meshes are within a certain area near the shape points, thus ensuring the accuracy of the elevation of the shape points calculated from the elevation of the point cloud meshes. If the number of point cloud meshes recalled at the initial recall level is equal to 0, the recall of meshes in a larger area will continue. The elevation of the surrounding point cloud can be used to calculate the elevation of the shape points in the area without point cloud, solving the problem that the elevation of some shape points cannot be calculated when there is no point cloud. At the same time, a preset level threshold is set to avoid the problem of inaccurate elevation calculation caused by the coverage area of ​​the recalled point cloud meshes being too large.

[0067] In one possible implementation, the step of filtering available point cloud meshes from the recalled point cloud meshes according to predetermined filtering rules includes:

[0068] From the recalled point cloud grids, select point cloud grids with more than a preset number of points within the grid and a difference between the elevation and the predicted elevation that is less than a preset difference as usable point cloud grids; wherein, the predicted elevation is the elevation of the shape points of the road element predicted based on the elevation information of existing elements stored in the parent database.

[0069] In this embodiment, the predetermined filtering rule is to select point cloud grids with relatively accurate elevations from the recalled point cloud grids as usable point cloud grids. If the number of point clouds in a grid is small, the elevation of the grid calculated based on the elevation information of the point clouds in that grid will be inaccurate. If the difference between the calculated elevation of the ground grid and the predicted elevation based on the existing information in the parent database is large, it indicates that the calculated elevation of the ground grid is also inaccurate. Therefore, the predetermined filtering rule can be to select point cloud grids with a number of point clouds in the grid exceeding a preset number and a difference between the elevation and the predicted elevation less than a preset difference.

[0070] In this embodiment, the predicted elevation can be the elevation of the shape point of the road element predicted based on the elevation information of existing elements stored in the parent database. If the road element is an existing element in the parent database, the elevation of the shape point of the road element stored in the parent database can be used as the predicted elevation of the shape point of the road element. If the road element is not an existing element in the parent database, the predicted elevation of the shape point of the road element can be determined based on the elevation information of other existing elements near the shape point of the road element stored in the parent database.

[0071] This embodiment can filter out point cloud grids with more accurate elevations from the recalled point cloud grids, where the number of point clouds in the grid exceeds a preset number and the difference between the elevation and the predicted elevation is less than a preset difference. This allows for the selection of usable point cloud grids, thereby making the elevation of shape points calculated from the elevation of the point cloud grids more accurate.

[0072] In one possible implementation, determining the elevation of the shape points of the road feature based on the elevation of the available point cloud mesh includes:

[0073] Obtain the location information of the road elements within the road;

[0074] In response to the orientation information of the road element in the road being located in the middle area of ​​the road, the elevation of the point cloud grid with the elevation sorted in a predetermined order in the available point cloud grid is determined as the elevation of the shape point of the road element.

[0075] In response to the orientation information of the road feature in the road being located in the edge region of the road, the elevation of the point cloud grid whose elevation is closest to the predicted elevation in the available point cloud grid is determined as the elevation of the shape point of the road feature.

[0076] In this embodiment, the orientation information of road elements in the road can be located in the middle area of ​​the road or in the edge area of ​​the road. For example, depending on the direction of travel, they can be located on the left or right side of the road. The orientation information of road elements in the road can be determined based on the travel trajectory when collecting point cloud data of road elements.

[0077] In this embodiment, if the orientation information of a road element within the road indicates it is located in the middle area of ​​the road, the available point cloud meshes can be first sorted by elevation. Then, the elevation of the point cloud mesh in the predetermined order can be determined as the elevation of the shape point of the road element. For example, the elevation of the point cloud mesh arranged from low to high at one-third of the number of available point cloud meshes can be determined as the elevation of the shape point of the road element. Assuming there are 9 available point cloud meshes, the elevation of the third point cloud mesh arranged from low to high can be determined as the elevation of the shape point of the road element.

[0078] In this embodiment, if the orientation information of the road element in the road is that it is located in the edge area of ​​the road, the elevation of the available point cloud grid whose elevation is closest to the predicted elevation can be determined as the elevation of the shape point of the road element.

[0079] In one possible implementation, obtaining the location information of the road element within the road includes:

[0080] Based on the trajectory data collected during the point cloud acquisition of the road elements, the initial orientation information of the road elements in the road is determined;

[0081] The initial orientation information is compared with the existing orientation information of the road elements stored in the parent database to obtain the comparison result;

[0082] If the comparison results are the same, the initial orientation information of the road element in the road is determined as the orientation information of the road element in the road.

[0083] In response to a difference in the comparison results, if the road element is located on a regular road, the orientation information of the road element on the road is determined to be located in the middle area of ​​the road; if the road element is located on a highway, the existing orientation information of the road element stored in the parent database is corrected based on the initial orientation information of the road element on the road, and the initial orientation information of the road element on the road is determined as the orientation information of the road element on the road.

[0084] In this embodiment, the initial orientation information of the road element in the road can be determined based on the trajectory data when collecting the point cloud of the road element. The initial orientation information can be located in the middle area, the left area, or the right area of ​​the road. For example, if the point cloud of the road element is located to the left of the trajectory route when collecting the point cloud of the road element, the initial orientation information is located in the left area of ​​the road. If the point cloud of the road element is located directly in front of the trajectory route when collecting the point cloud of the road element, the initial orientation information is located in the middle area of ​​the road.

[0085] In this embodiment, the predicted orientation information is the predicted orientation information of the road element in the road determined based on the orientation information of existing elements stored in the parent database. If the road element is an existing element in the parent database, the orientation information of the road element stored in the parent database can be used as the predicted orientation information of the road element. If the road element is not an existing element in the parent database, the predicted orientation information of the road element can be determined based on the orientation information of other existing elements near the road element stored in the parent database.

[0086] In this embodiment, the initial orientation information and the predicted orientation information are compared. If they are the same, the initial orientation information of the road element in the road can be directly determined as the orientation information of the road element in the road. If they are different, different processing schemes can be used according to different road types. If the road where the road element is located is a regular road, due to the complex road surface conditions of regular roads, it is impossible to determine whether the calculated initial orientation information is correct. In this case, the orientation information of the road element in the road can be determined as being located in the middle area of ​​the road. If the road where the road element is located is a highway, due to the relatively smooth road surface of highways, the calculated initial orientation information is usually correct. In this case, the existing orientation information of the road element stored in the master database can be modified to the orientation information of the road element in the road, and the initial orientation information of the road element in the road can be determined as the orientation information of the road element in the road.

[0087] In one possible implementation, the method further includes:

[0088] Based on the two-dimensional position information of the shape points of road elements with calculated elevations within the road area, the shape point grid corresponding to the road area is determined. The shape point grid is a grid of a predetermined shape divided according to the two-dimensional position.

[0089] The elevation of the shape point grid is determined based on the elevation of the shape points within the shape point grid.

[0090] Retrieve the shape point grid containing the shape point of the road element whose elevation has not been calculated, as well as the adjacent shape point grids, from the shape point grid corresponding to the road area;

[0091] For nearby shape points whose number of recalled shape point grids is greater than 0, the elevation of the nearby shape point is calculated based on the elevation of the shape point grid in which the nearby shape point is located and the elevation of the shape point grid adjacent to it.

[0092] In this embodiment, during the elevation calculation process described above, there may be instances of occlusion leading to the absence of point clouds, making it impossible to calculate the elevation of some shape points. In such cases, the grid corresponding to the road area, i.e., the shape point grid, can be redefined based on the two-dimensional position information of the shape points of the road elements whose elevations have already been calculated within the road area. This shape point grid is the same as the point cloud grid described above, or it can be a geohash grid. For example, GeoHash calculation can be performed on the two-dimensional position information of the shape points whose elevations have already been calculated to obtain the GeoHash grid code corresponding to each shape point. The grid corresponding to this grid code is the shape point grid where the point is located. In this way, the shape point grid corresponding to the shape points of the road elements whose elevations have already been calculated within the road area can be obtained, i.e., the shape point grid corresponding to the road area.

[0093] In this embodiment, to avoid the problem of some grids having no shape points due to the large distance between shape points, the density of shape points is increased before determining the shape point grid. For example, the density of shape points can be increased by 0.5m, and the elevation of the increased shape points is determined by the elevation of the adjacent shape points whose elevations have been calculated.

[0094] In this embodiment, for each shape point grid, the elevation of the shape point grid can be obtained by averaging the elevations of each shape point within the shape point grid.

[0095] In this embodiment, the shape point grids containing road element shape points whose elevations have not been calculated, as well as their adjacent shape point grids, can be recalled from the shape point grids corresponding to the road area. It should be noted that for shape points of road element shape points whose elevations have not been calculated, some shape points are located within or close to the corresponding shape point grid. When recalling the shape point grids corresponding to these shape points, the number of recalled shape point grids can be greater than 0; these shape points are referred to as nearby shape points. Other shape points are far from these shape point grids; when recalling the shape point grids corresponding to these shape points, the corresponding shape point grids cannot be recalled, i.e., the number of recalled shape point grids is equal to 0; these shape points are referred to as distant shape points.

[0096] In this embodiment, the two-dimensional location information of the nearby shape points of road elements within the road area whose elevations have not been calculated can be obtained, the elevations of the shape point grids where these nearby shape points are located and the shape point grids adjacent to them can be recalled, and the elevations of the nearby shape points can be calculated.

[0097] It should be noted that the method for calculating the elevation of the shape point to be determined is similar to the method for calculating the shape point elevation using point cloud elevation in the above embodiments. In response to the number of recalled shape point grids being greater than 0, usable shape point grids are selected from the recalled grids according to a predetermined filtering rule. The elevation of the nearby shape point is then calculated based on the combined elevations of the usable shape point grids. Here, it is not necessary to distinguish the orientation information of road elements; the elevation of the shape point grids whose elevations are ranked in a predetermined order among the usable shape point grids can be directly determined as the elevation of the nearby shape point. For example, the elevation of the shape point grids ranked from low to high at 1 / 3 of the number of usable shape point grids can be determined as the elevation of the nearby shape point.

[0098] This implementation can use the shape points of road elements whose elevations have already been calculated to redetermine the shape point grid corresponding to the road area. The shape point grid is then used to expand the scope of the recall, obtaining the shape point grid where the occluded shape point is located and the elevation of its adjacent shape point grids. Based on this, the elevation of the occluded shape points is calculated. In this way, the elevation of the occluded shape points is calculated using the elevation of the shape points of the road elements whose elevations have already been calculated, thus solving the problem that the elevation of some points cannot be calculated due to the absence of point clouds caused by occlusion.

[0099] In one possible implementation, the method further includes:

[0100] For distant shape points of road elements within the road area whose elevations have not been calculated, shape points whose distances from the distant shape points are within a predetermined distance range and whose elevations have been calculated are obtained as fitting shape points;

[0101] Based on the two-dimensional position information and elevation of the fitted shape points, a plane fitting is performed to obtain the fitted plane;

[0102] The fitting plane is extended to the distant shape point to obtain the elevation of the distant shape point.

[0103] In this embodiment, for shape points in occluded areas that are far apart, when recalling the corresponding shape point mesh, the corresponding shape point mesh cannot be recalled, that is, the number of recalled shape point meshes is equal to 0. Therefore, the elevation of these shape points cannot be calculated using the elevation of the shape point mesh. These shape points can be recorded as distant shape points. For distant shape points, a local plane fitting can be performed using the surrounding shape points whose elevations have been calculated to obtain a fitting plane. The fitting plane result is extended to the position of the distant shape point. The elevation of the fitting plane at the position of the distant shape point is the elevation of the distant shape point.

[0104] In this implementation, plane fitting refers to selecting the plane equation result that is closest to each fitted shape point based on the idea of ​​least squares. Figure 2B The diagram illustrates a plane fitting according to an embodiment of the present disclosure. The general expression for the plane equation is A*x + B*y + C*z + D = 0 (C ≠ 0), which can be denoted as... Then z = a0*x + a1*y + a2, for Figure 2B Given n (n≥3) shape points with calculated elevations, we need to fit a plane equation using the two-dimensional position information and elevation (i.e., three-dimensional coordinates) of these n shape points. The three-dimensional position information of the n shape points can be denoted as (x... i y i , z i ), i = 0, ..., n-1; for these n points (x i y i , z i ), as long as (a0*x+a1*y+a2-z) 2 The minimum value is given, where a0, a1, and a2 are unknown parameters.

[0105] Right now

[0106] The parameters a0, a1, and a2 can be used to obtain the result. Figure 2B The equation of the plane is shown, where C = 0 indicates that the plane is perpendicular to the ground, and the plane fitted by the shape point will never be perpendicular to the ground, therefore C ≠ 0.

[0107] This embodiment determines the elevation of distant shape points without point clouds by performing local plane fitting on shape points with calculated elevations and extending the fitted plane. It then uses other shape points with calculated elevations to determine distant shape points in occluded areas that are far apart in the absence of point clouds.

[0108] In one possible implementation, the method further includes:

[0109] The shape points of the road elements with calculated elevations are used for plane fitting to obtain the fitting plane for detection.

[0110] Calculate the point-to-surface distance from the shape point to the fitting plane used for detection;

[0111] Shape points whose point-to-surface distance is greater than a preset distance threshold are identified as abnormal jump points;

[0112] The elevation of the abnormal jump points is corrected based on the fitting plane used for detection.

[0113] In this embodiment, after calculating the elevation of the shape points of each road element within the road area using the above method, a plane fitting can be performed based on the three-dimensional position information of the shape points of the road elements with calculated elevations to obtain a fitting plane for detection. The point-to-plane distance from each shape point to the fitting plane for detection can be calculated. If the point-to-plane distance is too large, exceeding a preset distance threshold, it indicates that the position of the shape point deviates from other fitted shape points, which may be due to an abnormal elevation calculation of the shape point. In this case, the shape point can be identified as an abnormal jump point, and the abnormal jump point can be marked. The elevation of the abnormal jump point can be corrected based on the fitting plane for detection, so that the point-to-plane distance from the abnormal jump point to the fitting plane for detection is less than or equal to the preset distance threshold. For example, the elevation of the abnormal jump point can be directly corrected to the elevation of the fitting plane for detection at the two-dimensional position of the abnormal jump point.

[0114] This implementation can use the fitted plane obtained by plane fitting as a reference to determine abnormal jump points, and modify the abnormal jump points based on the fitted plane to make the calculated elevation of the shape points more accurate.

[0115] Figure 3A A flowchart illustrating a method for creating high-precision maps according to embodiments of the present disclosure is shown. Figure 3A As shown, the high-precision map production method includes the following steps S301-S303:

[0116] In step S301, two-dimensional road elements within the road area are generated based on the road data corresponding to the road area.

[0117] In step S302, the elevation of the shape points of the two-dimensional road element is calculated using the road element elevation determination method described above.

[0118] In step S303, three-dimensional road features within the road area are generated using the elevations of the two-dimensional road features and their shape points within the road area.

[0119] In one possible implementation, the high-precision map production method is applicable to computers, computing devices, servers, server clusters, and other devices capable of performing high-precision map production.

[0120] In one possible implementation, the road data includes trajectory data, image data, point cloud data, etc., collected by the data acquisition vehicle. The 2D production workshop can produce two-dimensional road elements within the road area based on the road data corresponding to the road area. The produced two-dimensional road elements include the two-dimensional shape information of the road elements. In the 3D production workshop, the elevation of the shape points of the road elements within the road area can be calculated according to the road element elevation determination method described in the above embodiment. The specific implementation process will not be detailed here. Using the two-dimensional road elements within the road area and the elevation of their shape points, three-dimensional road elements within the road area can be produced, thus realizing the production of three-dimensional road elements in high-precision maps.

[0121] This embodiment can determine the point cloud grid corresponding to the road area based on the point cloud data within the road area; and determine the elevation of the point cloud grid based on the elevation information of the point cloud within the point cloud grid. Then, based on the elevation of the point cloud grid where the shape point of the road element is located and the elevation of the point cloud grid adjacent to it, the elevation of the shape point of the road element is calculated. In this way, the 3D road element produced is more closely aligned with its point cloud, reducing manual work and thus improving production efficiency.

[0122] Figure 3B This diagram illustrates an application scenario of a method for determining road element elevation and creating high-precision maps according to an embodiment of this disclosure. For example... Figure 3B As shown, the data acquisition equipment mounted on the data acquisition vehicle 31 collects data such as trajectories, images, and point clouds of the road area and provides it to the map creation server 32. The map creation server 32 can first create the two-dimensional shapes of the road elements within the road area in a 2D creation workshop based on the collected data. Then, in a 3D creation workshop, it uses the aforementioned road element elevation determination method to determine the elevations of the shape points of the road elements within the road area. Based on these elevations, it produces three-dimensional road elements and updates the map data accordingly. This updated map data can be provided to the navigation server 33. The navigation server 33 can then provide navigation data to the location service terminal 34 based on this map data, enabling navigation, route planning, and other services.

[0123] Figure 4A A structural block diagram of a road feature elevation determination device according to an embodiment of the present disclosure is shown. This device can be implemented as part or all of an electronic device through software, hardware, or a combination of both. Figure 4A As shown, the road element elevation determination device includes:

[0124] The grid determination module 401 is configured to determine the point cloud grid corresponding to the road area based on the two-dimensional position information of the point cloud within the road area, wherein the point cloud grid is a grid of a predetermined shape divided according to the two-dimensional position.

[0125] The grid elevation determination module 402 is configured to determine the elevation of the point cloud grid based on the elevation information of the point cloud within the point cloud grid.

[0126] The shape point elevation determination module 403 is configured to calculate the elevation of the shape point of the road element based on the elevation of the point cloud grid where the shape point of the road element is located and the elevation of the point cloud grid adjacent to it within the road area.

[0127] In one possible implementation, the shape point elevation determination module 403 is configured as follows:

[0128] Recall the point cloud grid containing the shape points of the road element and the point cloud grid adjacent to it from the point cloud grid of the predetermined level corresponding to the road area;

[0129] In response to the fact that the number of recalled point cloud meshes is greater than 0, usable point cloud meshes are selected from the recalled point cloud meshes according to the predetermined filtering rules.

[0130] The elevation of the shape points of the road feature is determined based on the elevation of the available point cloud mesh.

[0131] In one possible implementation, the part of the point cloud grid containing the shape point of the road element and its adjacent point cloud grids that is retrieved from the point cloud grid of the predetermined level corresponding to the road area in the shape point elevation determination module 403 is configured as follows:

[0132] Recall the point cloud grid containing the shape points of the road element and the point cloud grid adjacent to them from the point cloud grid at the initial recall level corresponding to the road area;

[0133] In response to the number of recalled point cloud grids being 0, the point cloud grid containing the shape point of the road element and its adjacent point cloud grids are recalled from the point cloud grid of the previous level of the initial recall level, until the number of recalled point cloud grids is greater than 0 or the recall level reaches a preset level threshold, wherein the coverage area of ​​the previous level point cloud grid is larger than that of the point cloud grid of the initial recall level.

[0134] In one possible implementation, the part of the shape point elevation determination module 403 that filters available point cloud meshes from the recalled point cloud meshes according to a predetermined filtering rule is configured as follows:

[0135] From the recalled point cloud grids, select point cloud grids with more than a preset number of points within the grid and a difference between the elevation and the predicted elevation that is less than a preset difference as usable point cloud grids; wherein, the predicted elevation is the elevation of the shape points of the road element predicted based on the elevation information of existing elements stored in the parent database.

[0136] In one possible implementation, the portion of the shape point elevation determination module 403 that determines the elevation of the shape points of the road element based on the elevation of the available point cloud mesh is configured as follows:

[0137] Obtain the location information of the road elements within the road;

[0138] In response to the orientation information of the road element in the road being located in the middle area of ​​the road, the elevation of the point cloud grid with the elevation sorted in a predetermined order in the available point cloud grid is determined as the elevation of the shape point of the road element.

[0139] In response to the orientation information of the road feature in the road being located in the edge region of the road, the elevation of the point cloud grid whose elevation is closest to the predicted elevation in the available point cloud grid is determined as the elevation of the shape point of the road feature.

[0140] In one possible implementation, the portion of the shape point elevation determination module 403 that acquires the orientation information of the road element in the road is configured as follows:

[0141] Based on the trajectory data collected during the point cloud acquisition of the road elements, the initial orientation information of the road elements in the road is determined;

[0142] The initial orientation information and the predicted orientation information are compared to obtain a comparison result. The predicted orientation information is the predicted orientation information of the road element in the road, which is determined based on the orientation information of existing elements stored in the parent database.

[0143] If the comparison results are the same, the initial orientation information of the road element in the road is determined as the orientation information of the road element in the road.

[0144] In response to a difference in the comparison results, if the road element is located on a regular road, the orientation information of the road element on the road is determined to be located in the middle area of ​​the road; if the road element is located on a highway, the existing orientation information of the road element stored in the parent database is corrected based on the initial orientation information of the road element on the road, and the initial orientation information of the road element on the road is determined as the orientation information of the road element on the road.

[0145] In one possible implementation, the device further includes:

[0146] The shape point grid determination module is configured to determine the shape point grid corresponding to the road area based on the two-dimensional position information of the shape points of the road elements whose elevations have been calculated within the road area. The shape point grid is a grid of a predetermined shape divided according to the two-dimensional position.

[0147] The shape point grid elevation determination module is configured to determine the elevation of the shape point grid based on the elevation of the shape points within the shape point grid.

[0148] The proximity shape point elevation determination module is configured to recall the shape point grid containing the shape point of the road element whose elevation has not been calculated, as well as the adjacent shape point grids, from the shape point grid corresponding to the road area; for proximity shape points whose number of recalled shape point grids is greater than 0, the elevation of the proximity shape point is calculated based on the elevation of the shape point grid containing the proximity shape point and the adjacent shape point grids.

[0149] In one possible implementation, the device further includes:

[0150] The acquisition module is configured to retrieve distant shape points whose number of recalled shape point grids is equal to 0, and to acquire shape points of road elements whose distance from the distant shape points is within a predetermined distance range and whose elevation has been calculated as fitting shape points.

[0151] The plane fitting module is configured to perform plane fitting based on the two-dimensional position information and elevation of the fitted shape points to obtain a fitted plane.

[0152] A plane extension module is configured to extend the fitted plane to the distant shape point to obtain the elevation of the distant shape point.

[0153] In one possible implementation, the device further includes:

[0154] The fitting module is configured to perform plane fitting based on the shape points of road features with calculated elevations to obtain a fitting plane for detection.

[0155] The calculation module is configured to calculate the point-to-surface distance from the shape point to the fitting plane used for detection;

[0156] The jump point determination module is configured to identify shape points whose point-to-surface distance is greater than a preset distance threshold as abnormal jump points;

[0157] The correction module is configured to correct the elevation of the abnormal jump points based on the fitted plane used for detection.

[0158] Figure 4BA structural block diagram of a high-precision map creation apparatus according to an embodiment of the present disclosure is shown. This apparatus can be implemented as part or all of an electronic device through software, hardware, or a combination of both. Figure 4B As shown, the high-precision map production device includes:

[0159] The two-dimensional production module 404 is configured to produce two-dimensional road elements within the road area based on the original road data corresponding to the road area.

[0160] The elevation calculation module 405 is configured to calculate the elevation of the shape points of the two-dimensional road element using the road element elevation determination method described above.

[0161] The 3D production module 406 is configured to produce 3D road features within the road area using the elevations of 2D road features and their shape points within the road area.

[0162] The technical terms and features mentioned in this device implementation are the same or similar. For the explanation and description of the technical terms and features involved in this device, please refer to the explanation of the above method implementation, which will not be repeated here.

[0163] This disclosure also discloses an electronic device, Figure 5 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown.

[0164] like Figure 5 As shown, the electronic device 500 includes a memory 501 and a processor 502, wherein the memory 501 is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor 502 to implement the method according to embodiments of the present disclosure.

[0165] Figure 6 A schematic diagram of the structure of a computer system suitable for implementing the method according to embodiments of the present disclosure is shown.

[0166] like Figure 6 As shown, the computer system 600 includes a processing unit 601, which can execute various processes described in the above embodiments according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage portion 608 into a random access memory (RAM) 603. The RAM 603 also stores various programs and data required for the operation of the computer system 600. The processing unit 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0167] The following components are connected to I / O interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to I / O interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 610 as needed so that computer programs read from it can be installed into storage section 608 as needed. The processing unit 601 can be implemented as a CPU, GPU, TPU, FPGA, NPU, etc.

[0168] In particular, according to embodiments of this disclosure, the methods described above can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising computer instructions that, when executed by a processor, implement the steps of the methods described above. In such embodiments, the computer program product can be downloaded and installed from a network via communication section 609, and / or installed from removable media 611.

[0169] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0170] The units or modules described in the embodiments of this disclosure can be implemented in software or programmable hardware. The described units or modules can also be located in a processor, and the names of these units or modules do not necessarily constitute a limitation on the unit or module itself.

[0171] In another aspect, this disclosure also provides a computer-readable storage medium, which may be a computer-readable storage medium included in the electronic device or computer system described above; or it may be a standalone computer-readable storage medium not assembled into a device. The computer-readable storage medium stores one or more programs, which are used by one or more processors to perform the methods described in this disclosure.

[0172] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

Claims

1. A method for determining the elevation of road elements, comprising: Based on the two-dimensional position information of the point cloud within the road area, the point cloud grid corresponding to the road area is determined, wherein the point cloud grid is a grid of a predetermined shape divided according to the two-dimensional position. The elevation of the point cloud grid is determined based on the elevation information of the point cloud within the grid. Recall the point cloud grid containing the shape points of the road elements within the road area, as well as the point cloud grids adjacent to them, from the point cloud grid of the predetermined level corresponding to the road area; In response to the fact that the number of recalled point cloud meshes is greater than 0, available point cloud meshes are selected from the recalled point cloud meshes according to a predetermined filtering rule; and the elevation of the shape points of the road feature is determined based on the elevation of the available point cloud meshes.

2. The method of claim 1, wherein, in, The step of recalling the point cloud grid containing the shape points of the road element and its adjacent point cloud grids from the point cloud grid of the predetermined level corresponding to the road area includes: Recall the point cloud grid containing the shape points of the road element and the point cloud grid adjacent to them from the point cloud grid at the initial recall level corresponding to the road area; In response to the number of recalled point cloud grids being 0, the point cloud grid containing the shape point of the road element and its adjacent point cloud grids are recalled from the point cloud grid of the previous level of the initial recall level, until the number of recalled point cloud grids is greater than 0 or the recall level reaches a preset level threshold, wherein the coverage area of ​​the previous level point cloud grid is larger than that of the point cloud grid of the initial recall level.

3. The method according to claim 1, wherein, The step of filtering usable point cloud meshes from the recalled point cloud meshes according to predetermined filtering rules includes: From the recalled point cloud grids, select point cloud grids with more than a preset number of points within the grid and a difference between the elevation and the predicted elevation that is less than a preset difference as usable point cloud grids; wherein, the predicted elevation is the elevation of the shape points of the road element predicted based on the elevation information of existing elements stored in the parent database.

4. The method according to claim 3, wherein, Determining the elevation of the shape points of the road features based on the available point cloud mesh includes: Obtain the location information of the road elements within the road; In response to the orientation information of the road element in the road being located in the middle area of ​​the road, the elevation of the point cloud grid with the elevation sorted in a predetermined order in the available point cloud grid is determined as the elevation of the shape point of the road element. In response to the orientation information of the road feature in the road being located in the edge region of the road, the elevation of the point cloud grid whose elevation is closest to the predicted elevation in the available point cloud grid is determined as the elevation of the shape point of the road feature.

5. The method according to claim 4, wherein, The step of obtaining the location information of the road elements in the road includes: Based on the trajectory data collected during the point cloud acquisition of the road elements, the initial orientation information of the road elements in the road is determined; The initial orientation information and the predicted orientation information are compared to obtain a comparison result. The predicted orientation information is the predicted orientation information of the road element in the road, which is determined based on the orientation information of existing elements stored in the parent database. If the comparison results are the same, the initial orientation information of the road element in the road is determined as the orientation information of the road element in the road. In response to a difference in the comparison results, if the road element is located on a regular road, the orientation information of the road element on the road is determined to be located in the middle area of ​​the road; if the road element is located on a highway, the existing orientation information of the road element stored in the parent database is corrected based on the initial orientation information of the road element on the road, and the initial orientation information of the road element on the road is determined as the orientation information of the road element on the road.

6. The method according to claim 1, wherein, The method further includes: Based on the two-dimensional position information of the shape points of road elements with calculated elevations within the road area, the shape point grid corresponding to the road area is determined. The shape point grid is a grid of a predetermined shape divided according to the two-dimensional position. The elevation of the shape point grid is determined based on the elevation of the shape points within the shape point grid. Retrieve the shape point grid containing the shape point of the road element whose elevation has not been calculated, as well as the adjacent shape point grids, from the shape point grid corresponding to the road area; For nearby shape points whose number of recalled shape point grids is greater than 0, the elevation of the nearby shape point is calculated based on the elevation of the shape point grid in which the nearby shape point is located and the elevation of the shape point grid adjacent to it.

7. The method according to claim 6, wherein, The method further includes: For distant shape points where the number of recalled shape point grids is equal to 0, shape points of road elements whose distance from the distant shape points is within a predetermined distance range and whose elevations have been calculated are obtained as fitting shape points; Based on the two-dimensional position information and elevation of the fitted shape points, a plane fitting is performed to obtain the fitted plane; The fitting plane is extended to the distant shape point to obtain the elevation of the distant shape point.

8. The method according to any one of claims 1-7, wherein, The method further includes: The shape points of the road elements with calculated elevations are used for plane fitting to obtain the fitting plane for detection. Calculate the point-to-surface distance from the shape point to the fitting plane used for detection; Shape points whose point-to-surface distance is greater than a preset distance threshold are identified as abnormal jump points; The elevation of the abnormal jump points is corrected based on the fitting plane used for detection.

9. A method for producing high-precision maps, comprising: Based on the original road data corresponding to the road area, generate two-dimensional road elements within the road area; The elevation of the shape points of the two-dimensional road element is calculated using the method described in any one of claims 1 to 8; Using the elevations of two-dimensional road features and their shape points within the road area, three-dimensional road features within the road area are generated.

10. A road element elevation determination device, comprising: The grid determination module is configured to determine the point cloud grid corresponding to the road area based on the two-dimensional position information of the point cloud within the road area, wherein the point cloud grid is a grid of a predetermined shape divided according to the two-dimensional position. The grid elevation determination module is configured to determine the elevation of the point cloud grid based on the elevation information of the point cloud within the grid. The shape point elevation determination module is configured to recall the point cloud grid containing the shape points of the road elements within the road area and the point cloud grids adjacent to them from the point cloud grid of a predetermined level corresponding to the road area. In response to the fact that the number of recalled point cloud meshes is greater than 0, available point cloud meshes are selected from the recalled point cloud meshes according to a predetermined filtering rule; and the elevation of the shape points of the road feature is determined based on the elevation of the available point cloud meshes.

11. A high-precision map-making device, comprising: The two-dimensional production module is configured to produce two-dimensional road elements within the road area based on the original road data corresponding to the road area. The elevation calculation module is configured to calculate the elevation of the shape points of the two-dimensional road element using the method described in any one of claims 1 to 8; The 3D production module is configured to produce 3D road features within the road area using the elevations of 2D road features and their shape points within the road area.

12. An electronic device comprising a memory and a processor; wherein, The memory is used to store one or more computer instructions, which are executed by the processor to implement the steps of the method according to any one of claims 1 to 9.

13. A computer-readable storage medium having computer instructions stored thereon, wherein, When executed by a processor, the computer instructions implement the method described in any one of claims 1-9.

Images