Method, device, storage medium and equipment for generating three-dimensional high-precision vector map

Abstract

The present application discloses a method, device, storage medium and equipment for generating a three-dimensional high-precision vector map, which belongs to the field of unmanned driving technology. The method includes: generating a two-dimensional high-precision vector map; for each map element in the two-dimensional high-precision vector map, calculating the point cloud of the area of ​​interest of the two-dimensional coordinate points in the map element, the point cloud of the area of ​​interest is the three-dimensional point cloud data within the area of ​​interest of the two-dimensional coordinate points; calculating the elevation value of the two-dimensional coordinate points based on the point cloud of the area of ​​interest; generating a three-dimensional high-precision vector map based on each map element and the corresponding elevation value. The present application can generate a three-dimensional high-precision vector map based on post-elevation solution, which simplifies the map generation method, reduces the production cost of the map, and improves the map generation efficiency, which is conducive to the large-scale promotion and rapid update of the three-dimensional high-precision vector map. It can also generate a three-dimensional high-precision vector map based on the completed two-dimensional high-precision vector map, thereby improving the utilization value of the data.

Description

Technical Field

[0001] The present application relates to the field of unmanned driving technology, and in particular to a method, apparatus, storage medium and equipment for generating a three-dimensional high-precision vector map. Background Art

[0002] As an important component of the autonomous driving (unmanned driving) system, high-precision maps play an important role in the entire autonomous driving process. High-precision maps mainly include three parts: positioning layer, vector layer and dynamic layer. Among them, the vector layer mainly serves the perception and decision-making planning of unmanned vehicles. Figure 1 As shown in the figure, lane center lines, lane edges, stop lines, traffic lights, crosswalks and other elements are the core content of vector maps.

[0003] Based on the mapping needs of autonomous vehicles, existing high-precision vector maps are available in two formats: two-dimensional and three-dimensional. Two-dimensional maps assume the ground is horizontal, creating a map representation in the X and Y dimensions by orthographically projecting real data along the Z axis. Based on the two-dimensional map, the true elevation of each ground location is calculated to create a three-dimensional high-precision vector map represented in the X, Y, and Z dimensions.

[0004] as follows Figure 2 As shown in the figure, existing 3D high-precision vector maps are mainly produced based on professional 3D interactive platforms. The production process mainly includes:

[0005] (1) Obtain three-dimensional dense point cloud data in the map reference coordinate system;

[0006] (2) Capture the 3D key point set of a specific map element by controlling 6 degrees of freedom (X / Y / Z translation vectors and P / O / K rotation vectors) to perform 3D interaction;

[0007] (3) Use the specific tools provided by the system to convert the three-dimensional key point set into three-dimensional map elements, and edit and maintain the topological relationship between the three-dimensional map elements to generate a complete vector topological map, and then you can get a three-dimensional high-precision vector map.

[0008] Since three-dimensional interaction requires the control of three translations and three rotations, a total of six degrees of freedom, its interaction complexity is higher and its interaction efficiency is lower than that of two-dimensional interaction. Therefore, the production cost and mapping cycle of three-dimensional high-precision vector maps based on three-dimensional interactive environments have remained high, and the requirements and thresholds for cartographers and mapping equipment are higher, which is not conducive to the large-scale promotion and rapid updating of three-dimensional high-precision vector maps. Summary of the Invention

[0009] This application provides a method, apparatus, storage medium, and device for generating three-dimensional high-precision vector maps. These methods address the high production costs and long mapping cycles associated with generating three-dimensional high-precision vector maps based on three-dimensional interaction, hindering their widespread promotion and rapid updating. The technical solution is as follows:

[0010] In one aspect, a method for generating a three-dimensional high-precision vector map is provided, the method comprising:

[0011] Generate two-dimensional high-precision vector maps;

[0012] For each map element in the two-dimensional high-precision vector map, calculating an area of ​​interest point cloud of the two-dimensional coordinate point in the map element, the area of ​​interest point cloud being three-dimensional point cloud data within the area of ​​interest of the two-dimensional coordinate point; and calculating an elevation value of the two-dimensional coordinate point based on the area of ​​interest point cloud;

[0013] Generate a three-dimensional high-precision vector map based on each map element and the corresponding elevation value.

[0014] In a possible implementation, computing the elevation value of the two-dimensional coordinate point based on the point cloud of the area of ​​interest includes:

[0015] Acquire the elevation value of each three-dimensional point cloud data from the point cloud of the area of ​​interest to obtain an elevation value set;

[0016] Calculating a central elevation value according to the distribution of each elevation value in the elevation value set;

[0017] The center elevation value is used as the elevation value of the two-dimensional coordinate point.

[0018] In a possible implementation, calculating the central elevation value according to the distribution of each elevation value in the elevation value set includes:

[0019] Input each elevation value in the elevation value set into a calculation formula until the calculation formula outputs two consecutive values ​​that meet a preset condition, and use the last output value as the center elevation value;

[0020] The calculation formula is Among them, x j represents the output value of the jth operation, and x 0 is the minimum value in the elevation value set, x i represents the i-th elevation value in the elevation value set, Indicates x j-1 As the center, with h as the neighborhood radius, g(●) represents x i The corresponding weight kernel function.

[0021] In one possible implementation, generating a two-dimensional high-precision vector map includes:

[0022] Get the original point cloud data;

[0023] Converting the original point cloud data into a two-dimensional image using a point cloud projection image algorithm;

[0024] The two-dimensional high-precision vector map is generated according to the two-dimensional image.

[0025] In a possible implementation, calculating the area of ​​interest point cloud of the two-dimensional coordinate points in the map elements includes:

[0026] For each two-dimensional coordinate point in the map element, creating an area of ​​interest of the two-dimensional coordinate point with the coordinate of the two-dimensional coordinate point as the center and a predetermined value as the radius;

[0027] Filtering three-dimensional point cloud data located within the region of interest from the original point cloud data;

[0028] The filtered three-dimensional point cloud data is used to form a point cloud of the region of interest of the two-dimensional coordinate points.

[0029] In one aspect, a device for generating a three-dimensional high-precision vector map is provided, the device comprising:

[0030] Two-dimensional map generation module, used to generate two-dimensional high-precision vector maps;

[0031] An elevation value calculation module is configured to calculate, for each map element in the two-dimensional high-precision vector map, an area of ​​interest point cloud of a two-dimensional coordinate point in the map element, the area of ​​interest point cloud being three-dimensional point cloud data within an area of ​​interest of the two-dimensional coordinate point; and calculate an elevation value of the two-dimensional coordinate point based on the area of ​​interest point cloud;

[0032] The three-dimensional map generation module is used to generate a three-dimensional high-precision vector map based on each map element and the corresponding elevation value.

[0033] In a possible implementation, the elevation value calculation module is further configured to:

[0034] Acquire the elevation value of each three-dimensional point cloud data from the point cloud of the area of ​​interest to obtain an elevation value set;

[0035] Calculating a central elevation value according to the distribution of each elevation value in the elevation value set;

[0036] The center elevation value is used as the elevation value of the two-dimensional coordinate point.

[0037] In a possible implementation, the elevation value calculation module is further configured to:

[0038] Input each elevation value in the elevation value set into a calculation formula until the calculation formula outputs two consecutive values ​​that meet a preset condition, and use the last output value as the center elevation value;

[0039] The calculation formula is Among them, x j represents the output value of the jth operation, and x 0 is the minimum value in the elevation value set, x i represents the i-th elevation value in the elevation value set, Indicates x j-1 As the center, with h as the neighborhood radius, g(●) represents x i The corresponding weight kernel function.

[0040] On the one hand, a computer-readable storage medium is provided, wherein the storage medium stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement the method for generating a three-dimensional high-precision vector map as described above.

[0041] On the one hand, a computer device is provided, which includes a processor and a memory, wherein the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method for generating a three-dimensional high-precision vector map as described above.

[0042] The beneficial effects of the technical solution provided by this application include at least:

[0043] By generating a 2D high-precision vector map, for each map feature in the 2D high-precision vector map, a point cloud of the area of ​​interest (ROI) can be calculated for the 2D coordinates of the map feature. The elevation values ​​of the 2D coordinates of the ROI point cloud are then calculated based on the ROI point cloud. Finally, a 3D high-precision vector map is generated based on each map feature and its corresponding elevation value. This simplifies map generation, reduces production costs, and improves efficiency, facilitating the widespread promotion and rapid updating of 3D high-precision vector maps. Furthermore, 3D high-precision vector maps can be generated based on pre-drawn 2D high-precision vector maps, increasing the value of the data. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0045] Figure 1 This is a schematic diagram of a high-precision vector map shown in the prior art;

[0046] Figure 2 It is a schematic diagram of a three-dimensional interactive platform shown in the prior art;

[0047] Figure 3 This is a flow chart of a method for generating a three-dimensional high-precision vector map provided by one embodiment of the present application;

[0048] Figure 4 This is a flowchart of a method for generating a three-dimensional high-precision vector map provided by another embodiment of the present application;

[0049] Figure 5 is a schematic diagram of a point cloud of an area of ​​interest provided by another embodiment of the present application;

[0050] Figure 6 This is a flowchart of a method for generating a three-dimensional high-precision vector map provided by another embodiment of the present application;

[0051] Figure 7 This is a structural block diagram of a device for generating a three-dimensional high-precision vector map provided in yet another embodiment of the present application. DETAILED DESCRIPTION

[0052] In order to make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0053] Please refer to Figure 3 , which shows a method flow chart of a method for generating a three-dimensional high-precision vector map provided by an embodiment of the present application. The method for generating a three-dimensional high-precision vector map can be applied to a computer device. The method for generating a three-dimensional high-precision vector map can include:

[0054] Step 301: Generate a two-dimensional high-precision vector map.

[0055] The computer device can generate a two-dimensional high-precision vector map based on the current mature two-dimensional vector mapping technology. The specific generation method is not limited in this embodiment.

[0056] Step 302 , for each map element in the two-dimensional high-precision vector map, calculate an area of ​​interest point cloud of the two-dimensional coordinate points in the map element, wherein the area of ​​interest point cloud is three-dimensional point cloud data within the area of ​​interest of the two-dimensional coordinate points; and calculate the elevation value of the two-dimensional coordinate points based on the area of ​​interest point cloud.

[0057] In a 2D high-precision vector map, map elements are stored in three geometric forms: 2D points, lines, and surfaces. For point-based map elements, each map element is represented by a single 2D coordinate point. For linear and surface map elements, each map element is represented by one or more 2D coordinate point strings. Computer equipment can collect 2D point or 2D point string information within the point, line, and surface map elements in a 2D high-precision vector map to obtain a 2D coordinate point set, and then convert each 2D coordinate point in the 2D coordinate point set into a 3D coordinate point.

[0058] Specifically, for each two-dimensional coordinate point, the computer device first obtains its corresponding area of ​​interest point cloud, and then calculates its corresponding elevation value based on the area of ​​interest point cloud. The specific calculation method is described in detail below.

[0059] Step 303: Generate a three-dimensional high-precision vector map based on each map element and the corresponding elevation value.

[0060] For each 2D coordinate point in a map feature, the computer device obtains the elevation value of the point cloud of the area of ​​interest corresponding to the 2D coordinate point and combines the elevation value with the 2D coordinate point to obtain the 3D coordinate point. For example, if the original 2D coordinate point is (X, Y), the elevation value is used as the Z-axis value and combined with it to obtain the 3D coordinate (X, Y, Z).

[0061] After obtaining the three-dimensional coordinate points corresponding to all the two-dimensional coordinate points, the computer device can store the three-dimensional coordinate points according to the three-dimensional map data specifications to obtain a three-dimensional high-precision vector map.

[0062] It should be noted that when a three-dimensional high-precision vector map needs to be displayed, the computer device also needs to draw and render the three-dimensional coordinate points.

[0063] In summary, the method for generating a three-dimensional high-precision vector map provided in the embodiments of the present application generates a two-dimensional high-precision vector map. For each map element in the two-dimensional high-precision vector map, the point cloud of the area of ​​interest of the two-dimensional coordinate points in the map element can be calculated. The elevation values ​​of the two-dimensional coordinate points are then calculated based on the point cloud of the area of ​​interest. Finally, a three-dimensional high-precision vector map is generated based on each map element and the corresponding elevation value. In this way, a three-dimensional high-precision vector map can be generated based on elevation post-solution, which simplifies the map generation method, reduces the production cost of the map, and improves the map generation efficiency, which is conducive to the large-scale promotion and rapid updating of three-dimensional high-precision vector maps. In addition, a three-dimensional high-precision vector map can be generated based on an already drawn two-dimensional high-precision vector map, thereby increasing the utilization value of the data.

[0064] Please refer to Figure 4, which shows a method flow chart of a method for generating a three-dimensional high-precision vector map provided by an embodiment of the present application. The method for generating a three-dimensional high-precision vector map can be applied to a computer device. The method for generating a three-dimensional high-precision vector map can include:

[0065] Step 401: Obtain original point cloud data.

[0066] The computer device can obtain three-dimensional original point cloud data. The original point cloud data here can be obtained by shooting the computer device or obtained from the network or other devices. The source of the original point cloud data is not limited in this embodiment.

[0067] Step 402: Convert the original point cloud data into a two-dimensional image using a point cloud projection image algorithm.

[0068] The computer device can use the currently mature point cloud projection image algorithm to convert the three-dimensional original point cloud data into a two-dimensional image. The specific projection method is not limited in this embodiment.

[0069] Step 403: Generate a two-dimensional high-precision vector map based on the two-dimensional image.

[0070] The computer device can use the current mature mapping technology to generate a two-dimensional high-precision vector map of uniform specifications from a two-dimensional image. The specific generation method is not limited in this embodiment.

[0071] Step 404 , for each map element in the two-dimensional high-precision vector map, calculate an area of ​​interest point cloud of the two-dimensional coordinate points in the map element, wherein the area of ​​interest point cloud is three-dimensional point cloud data within the area of ​​interest of the two-dimensional coordinate points.

[0072] In a 2D high-precision vector map, map elements are stored in three geometric forms: 2D points, lines, and surfaces. For point-based map elements, each map element is represented by a single 2D coordinate point. For linear and surface map elements, each map element is represented by one or more 2D coordinate point strings. Computer equipment can collect 2D point or 2D point string information within the point, line, and surface map elements in a 2D high-precision vector map to obtain a 2D coordinate point set, and then convert each 2D coordinate point in the 2D coordinate point set into a 3D coordinate point.

[0073] Specifically, calculating the point cloud of the area of ​​interest of the two-dimensional coordinate points in the map elements may include: for each two-dimensional coordinate point in the map element, creating an area of ​​interest of the two-dimensional coordinate point with the coordinates of the two-dimensional coordinate point as the center and a predetermined value as the radius; filtering the three-dimensional point cloud data located in the area of ​​interest from the original point cloud data; and composing the filtered three-dimensional point cloud data into the point cloud of the area of ​​interest of the two-dimensional coordinate point.

[0074] Assuming the two-dimensional coordinates of the two-dimensional coordinate point are (X, Y) and the radius is r, the computer device can create the following Figure 5 The region of interest (ROI) is shown, and the three-dimensional point cloud data (the points in the circular area) located in the region of interest in the original point cloud data are combined into a point cloud of the region of interest.

[0075] Step 405 : Obtain the elevation value of each three-dimensional point cloud data from the point cloud of the area of ​​interest to obtain an elevation value set.

[0076] The three-dimensional point cloud data contains elevation values, which can be directly obtained by computer equipment.

[0077] Assuming that the total number of three-dimensional point cloud data contained in the point cloud of the area of ​​interest is n, the elevation value in each three-dimensional point cloud data is taken out to obtain a one-dimensional space point set X containing n samples. The one-dimensional space point set X is the elevation value set X.

[0078] Step 406: Calculate the central elevation value based on the distribution of each elevation value in the elevation value set.

[0079] Specifically, calculating the central elevation value according to the distribution of each elevation value in the elevation value set may include: inputting each elevation value in the elevation value set into a calculation formula, stopping after two consecutive output values ​​of the calculation formula meet a preset condition, and using the last output value as the central elevation value;

[0080] The calculation formula is Among them, x j represents the output value of the jth operation, and x 0 is the minimum value in the elevation value set, x i represents the i-th elevation value in the elevation value set, Indicates x j -1 As the center, with h as the neighborhood radius, g(●) represents x i The corresponding weight kernel function.

[0081] Where g(●) is x i The corresponding weight kernel function, the weight kernel function mentioned here may include but is not limited to a Gaussian kernel function, a mean kernel function and other types of custom kernel functions.

[0082] When j = 0, the computer device can traverse the elevation value set X and find the minimum value, which is recorded as x 0 When j = 1, x j-1 =x 0 =0, then according to x i 、 and h to calculate x1 ; When j = 2, x j-1 =x 1 , and then according to x i 、 and h to calculate x 2 , and so on, until the change in x calculated twice before and after is less than the error threshold or j reaches the specified number of iterations, and the last output x is recorded as x * , then x * is the final center elevation value.

[0083] Step 407: Use the center elevation value as the elevation value of the two-dimensional coordinate point.

[0084] Step 408: Generate a three-dimensional high-precision vector map based on each map element and the corresponding elevation value.

[0085] For each 2D coordinate point in a map feature, the computer device obtains the elevation value of the point cloud of the area of ​​interest corresponding to the 2D coordinate point and combines the elevation value with the 2D coordinate point to obtain the 3D coordinate point. For example, if the original 2D coordinate point is (X, Y), the elevation value is used as the Z-axis value and combined with it to obtain the 3D coordinate (X, Y, Z).

[0086] After obtaining the three-dimensional coordinate points corresponding to all the two-dimensional coordinate points, the computer device can store the three-dimensional coordinate points according to the three-dimensional map data specifications to obtain a three-dimensional high-precision vector map.

[0087] It should be noted that when a three-dimensional high-precision vector map needs to be displayed, the computer device also needs to draw and render the three-dimensional coordinate points.

[0088] In simple terms, the computer equipment first collects 3D point clouds, performs 2D projection and 2D vectoring on the 3D point clouds to obtain a 2D high-precision vector map, then collects 2D points from the 2D high-precision vector map, extracts ROI point clouds from the 3D point clouds based on the 2D points, and then performs elevation calculation on the ROI point clouds. Finally, a 3D high-precision vector map is generated based on the calculated elevation values. The generation process is as follows: Figure 6 shown.

[0089] In summary, the method for generating a three-dimensional high-precision vector map provided in the embodiments of the present application generates a two-dimensional high-precision vector map. For each map element in the two-dimensional high-precision vector map, the point cloud of the area of ​​interest of the two-dimensional coordinate points in the map element can be calculated. The elevation values ​​of the two-dimensional coordinate points are then calculated based on the point cloud of the area of ​​interest. Finally, a three-dimensional high-precision vector map is generated based on each map element and the corresponding elevation value. In this way, a three-dimensional high-precision vector map can be generated based on elevation post-solution, which simplifies the map generation method, reduces the production cost of the map, and improves the map generation efficiency, which is conducive to the large-scale promotion and rapid updating of three-dimensional high-precision vector maps. In addition, a three-dimensional high-precision vector map can be generated based on an already drawn two-dimensional high-precision vector map, thereby increasing the utilization value of the data.

[0090] Please refer to Figure 7 , which shows a structural block diagram of a device for generating a three-dimensional high-precision vector map provided by one embodiment of the present application. The device for generating a three-dimensional high-precision vector map can be applied to a computer device. The device for generating a three-dimensional high-precision vector map can include:

[0091] A two-dimensional map generation module 710 is used to generate a two-dimensional high-precision vector map;

[0092] The elevation value calculation module 720 is configured to calculate, for each map element in the two-dimensional high-precision vector map, an area of ​​interest point cloud of the two-dimensional coordinate points in the map element, wherein the area of ​​interest point cloud is three-dimensional point cloud data within the area of ​​interest of the two-dimensional coordinate points; and calculate the elevation value of the two-dimensional coordinate points based on the area of ​​interest point cloud;

[0093] The three-dimensional map generation module 730 is used to generate a three-dimensional high-precision vector map according to each map element and the corresponding elevation value.

[0094] In an optional embodiment, the elevation value calculation module 720 is further configured to:

[0095] Obtaining the elevation value of each three-dimensional point cloud data from the point cloud of the area of ​​interest to obtain an elevation value set;

[0096] Calculate the central elevation value according to the distribution of each elevation value in the elevation value set;

[0097] Use the center elevation value as the elevation value of the two-dimensional coordinate point.

[0098] In an optional embodiment, the elevation value calculation module 720 is further configured to:

[0099] Each elevation value in the elevation value set is input into the calculation formula until the calculation formula outputs two consecutive values ​​that meet the preset conditions, and the last output value is used as the center elevation value;

[0100] The calculation formula is Among them, x j represents the output value of the jth operation, and x 0 is the minimum value in the elevation value set, x i represents the i-th elevation value in the elevation value set, Indicates x j -1 As the center, with h as the neighborhood radius, g(●) represents x i The corresponding weight kernel function.

[0101] In an optional embodiment, the two-dimensional map generation module 710 is further configured to:

[0102] Get the original point cloud data;

[0103] Use point cloud projection image algorithm to convert the original point cloud data into a two-dimensional image;

[0104] Generate a two-dimensional high-precision vector map based on a two-dimensional image.

[0105] In an optional embodiment, the elevation value calculation module 720 is further configured to:

[0106] For each two-dimensional coordinate point in the map element, a region of interest of the two-dimensional coordinate point is created with the coordinate of the two-dimensional coordinate point as the center and a predetermined value as the radius;

[0107] Filtering 3D point cloud data within the region of interest from the original point cloud data;

[0108] The filtered three-dimensional point cloud data is composed of a point cloud of the region of interest with two-dimensional coordinate points.

[0109] In summary, the device for generating a three-dimensional high-precision vector map provided in the embodiments of the present application generates a two-dimensional high-precision vector map. For each map element in the two-dimensional high-precision vector map, the device can calculate the area of ​​interest point cloud of the two-dimensional coordinate points in the map element, then calculate the elevation values ​​of the two-dimensional coordinate points based on the area of ​​interest point cloud. Finally, a three-dimensional high-precision vector map is generated based on each map element and the corresponding elevation value. In this way, a three-dimensional high-precision vector map can be generated based on elevation post-solution, which simplifies the map generation method, reduces the production cost of the map, and improves the map generation efficiency, which is conducive to the large-scale promotion and rapid updating of three-dimensional high-precision vector maps. In addition, a three-dimensional high-precision vector map can be generated based on an already drawn two-dimensional high-precision vector map, thereby increasing the utilization value of the data.

[0110] One embodiment of the present application provides a computer-readable storage medium, wherein the storage medium stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement the method for generating a three-dimensional high-precision vector map as described above.

[0111] An embodiment of the present application provides a computer device, which includes a processor and a memory, wherein the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method for generating a three-dimensional high-precision vector map as described above.

[0112] It should be noted that the above-described embodiments of the apparatus for generating a 3D high-precision vector map illustrate the division of the functional modules described above only as an example. In actual applications, the above-described functions can be assigned to different functional modules as needed, i.e., the internal structure of the apparatus for generating a 3D high-precision vector map can be divided into different functional modules to perform all or part of the functions described above. Furthermore, the apparatus for generating a 3D high-precision vector map provided in the above-described embodiments shares the same concept as the embodiment of the method for generating a 3D high-precision vector map. The detailed implementation process is described in the method embodiment and will not be further elaborated here.

[0113] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or by a program to instruct the relevant hardware, and the program may be stored in a computer-readable storage medium, which may be a read-only memory, a disk, or an optical disk, etc.

[0114] The above description is not intended to limit the embodiments of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of the embodiments of the present application.

Claims

1. A method for generating a three-dimensional high-precision vector map, characterized in that: The method comprises: Generate two-dimensional high-precision vector maps; For each map element in the two-dimensional high-precision vector map, calculating an area of ​​interest point cloud of the two-dimensional coordinate point in the map element, the area of ​​interest point cloud being three-dimensional point cloud data within the area of ​​interest of the two-dimensional coordinate point; and calculating an elevation value of the two-dimensional coordinate point based on the area of ​​interest point cloud; Generate a three-dimensional high-precision vector map based on each map element and its corresponding elevation value; Calculating the elevation value of the two-dimensional coordinate point based on the point cloud of the area of ​​interest includes: obtaining the elevation value of each three-dimensional point cloud data from the point cloud of the area of ​​interest to obtain an elevation value set; calculating a center elevation value based on the distribution of each elevation value in the elevation value set; and using the center elevation value as the elevation value of the two-dimensional coordinate point; The calculation of the central elevation value according to the distribution of each elevation value in the elevation value set includes: inputting each elevation value in the elevation value set into a calculation formula until the output value of the calculation formula meets the preset condition twice in a row, and taking the last output value as the central elevation value; the calculation formula is Among them, x j represents the output value of the jth operation, and x 0 is the minimum value in the elevation value set, x i represents the i-th elevation value in the elevation value set, Indicates x j-1 As the center, with h as the neighborhood radius, g(●) represents x i The corresponding weight kernel function.

2. The method according to claim 1, characterized in that The generating of a two-dimensional high-precision vector map includes: Get original point cloud data; Converting the original point cloud data into a two-dimensional image using a point cloud projection image algorithm; The two-dimensional high-precision vector map is generated according to the two-dimensional image.

3. The method according to claim 1 or 2, characterized in that The calculating of the point cloud of the area of ​​interest of the two-dimensional coordinate points in the map elements includes: For each two-dimensional coordinate point in the map element, creating an area of ​​interest of the two-dimensional coordinate point with the coordinate of the two-dimensional coordinate point as the center and a predetermined value as the radius; Filtering three-dimensional point cloud data located within the region of interest from the original point cloud data; The filtered three-dimensional point cloud data is used to form a point cloud of the region of interest of the two-dimensional coordinate points.

4. A device for generating a three-dimensional high-precision vector map, characterized in that: The device comprises: Two-dimensional map generation module, used to generate two-dimensional high-precision vector maps; An elevation value calculation module is configured to calculate, for each map element in the two-dimensional high-precision vector map, an area of ​​interest point cloud of a two-dimensional coordinate point in the map element, the area of ​​interest point cloud being three-dimensional point cloud data within an area of ​​interest of the two-dimensional coordinate point; and calculate an elevation value of the two-dimensional coordinate point based on the area of ​​interest point cloud; A three-dimensional map generation module is used to generate a three-dimensional high-precision vector map based on each map element and its corresponding elevation value; The elevation value calculation module is further configured to: obtain the elevation value of each three-dimensional point cloud data from the point cloud of the area of ​​interest to obtain an elevation value set; calculate the center elevation value based on the distribution of each elevation value in the elevation value set; and use the center elevation value as the elevation value of the two-dimensional coordinate point; The elevation value calculation module is further used to: input each elevation value in the elevation value set into a calculation formula until the calculation formula outputs two consecutive values ​​that meet the preset conditions, and then stop, and use the last output value as the center elevation value; the calculation formula is Among them, x j represents the output value of the jth operation, and x 0 is the minimum value in the elevation value set, x i represents the i-th elevation value in the elevation value set, Indicates x j-1 As the center, with h as the neighborhood radius, g(●) represents x i The corresponding weight kernel function.

5. A computer-readable storage medium, characterized in that The storage medium stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the method for generating a three-dimensional high-precision vector map as described in any one of claims 1 to 3.

6. A computer device, characterized in that: The computer device includes a processor and a memory, wherein the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method for generating a three-dimensional high-precision vector map as described in any one of claims 1 to 3.

Images