Method and device for determining waypoint height and storage medium

Abstract

The invention discloses a method and equipment for determining the height of a waypoint and a storage medium, and belongs to the technical field of waypoint measurement and calculation. The method comprises the following steps: acquiring a gridding area elevation difference data set of a target area, and target longitude and latitude and target altitude of a target waypoint in the target area; determining a target cell of the target waypoint in the plurality of cells according to the target latitude and longitude and the gridding area elevation difference data set; based on a pre-trained neural network elevation difference prediction model, obtaining a predicted elevation difference of the target waypoint according to the target latitude and longitude and the cell vertex data corresponding to the target cell; and according to the predicted elevation difference and the target altitude, determining the ellipsoid height of the target waypoint to obtain the waypoint height of the target waypoint. According to the scheme, the accuracy of determining the waypoint height is improved.

Description

Technical Field

[0001] The present application relates to the technical field of waypoint measurement, and in particular to a method, device and storage medium for determining the altitude of a waypoint. Background Art

[0002] To create a route for a drone, the user needs to set waypoints in the map. Each waypoint has a waypoint longitude, latitude, and altitude. The existing waypoint altitude confirmation method is to divide the map into a certain number of squares, take a point within each square, and calculate and process a preliminary waypoint altitude based on parameters such as the altitude of the point. If the waypoint location selected by the user is located at a point within a square, the waypoint altitude of the waypoint selected by the user is the preliminary waypoint altitude of the above point. That is, the waypoint altitude at any position within each square is the same. Therefore, the existing waypoint altitude confirmation method has the technical problem of low accuracy. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a method, device and storage medium for determining waypoint altitude, so as to solve the technical problem that the waypoint altitude confirmation method in the prior art has low accuracy.

[0004] In order to achieve the above-mentioned object, the first aspect of the present application provides a method for determining a waypoint altitude, comprising:

[0005] Obtaining a gridded regional elevation difference data set of a target area, a target longitude and latitude of a target waypoint in the target area, and a target altitude; wherein the gridded regional elevation difference data set includes: cell vertex data corresponding to a plurality of cells, the cell vertex data includes vertex longitude and latitude data of each cell and actual vertex elevation difference data;

[0006] Determine a target cell of a target waypoint among a plurality of cells according to the target longitude and latitude and the gridded regional elevation difference data set;

[0007] Based on the pre-trained neural network elevation difference prediction model, the predicted elevation difference of the target waypoint is obtained according to the target longitude and latitude and the cell vertex data corresponding to the target cell;

[0008] According to the predicted elevation difference and the target altitude, the ellipsoid height of the target waypoint is determined to obtain the waypoint height of the target waypoint.

[0009] In some embodiments of the present application, before obtaining the gridded regional elevation difference dataset, the method also includes: gridding the target area to generate a longitude and latitude grid of the target area; wherein the longitude and latitude grid includes a plurality of cells, each cell having a plurality of cell vertices containing longitude and latitude data; obtaining actual elevation difference data of each cell vertex based on the longitude and latitude data of each cell vertex; generating a gridded regional elevation difference dataset based on the longitude and latitude data of each cell vertex and the actual elevation difference data.

[0010] In some embodiments of the present application, after generating the latitude and longitude grid of the target area, the method also includes: selecting an expected point in each cell; determining the expected point longitude and latitude data and the actual elevation difference data of each expected point; generating a training vector for each expected point based on the expected point and the longitude and latitude grid, wherein the training vector includes: the expected point longitude and latitude data corresponding to the expected point, and the vertex longitude and latitude data and actual elevation difference data of multiple cell vertices of the cell where the expected point is located; training a preset neural network model based on the training vector so that the predicted elevation difference output by the preset neural network model approaches the actual elevation difference data of the expected point until the model converges to obtain a neural network elevation difference prediction model.

[0011] In some embodiments of the present application, cells in the latitude and longitude grid are tileable regular polygon cells.

[0012] In some embodiments of the present application, the cells in the latitude and longitude grid are tileable regular polygonal cells; the desired point is the center of the cell.

[0013] In some embodiments of the present application, regular polygonal cells may be tiled into square cells.

[0014] In some embodiments of the present application, there are multiple training vectors, and the multiple training vectors are divided into a training set, a validation set, and a test set; wherein the training set has the largest number of training vectors.

[0015] In some embodiments of the present application, the cell where the target waypoint is located is determined based on the target longitude and latitude and in a gridded area elevation difference dataset, including: determining multiple target cell vertex data with the smallest difference from the target longitude and latitude based on the target longitude and latitude and the gridded area elevation difference dataset; determining the target cell based on the multiple target cell vertex data.

[0016] A second aspect of the present application provides a waypoint altitude determination device, comprising: a memory configured to store instructions; and a processor configured to call the instructions from the memory and implement the method for waypoint altitude determination provided in accordance with any one of the above embodiments when executing the instructions.

[0017] A third aspect of the present application provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute the method for determining the waypoint altitude provided according to any of the above embodiments.

[0018] Through the above technical solution, after obtaining the target longitude and latitude and the target altitude of the target waypoint, the target cell where the target waypoint is located and the vertex longitude and latitude data of multiple cell vertices of the target cell and the actual elevation difference data of the vertex are determined in the gridded area elevation difference data set according to the target longitude and latitude. The target longitude and latitude, the vertex longitude and latitude data of multiple cell vertices and the actual elevation difference data of the vertex are input into the neural network elevation difference prediction model, and the model outputs the predicted elevation difference of the target waypoint accordingly. Then, the ellipsoid height of the target waypoint can be determined according to the predicted elevation difference and the target altitude. Since the terrain of the actual target area is complex and does not change linearly, in order to achieve better elevation difference prediction results, the neural network elevation difference prediction model is used, and the elevation difference of any waypoint in the target area is predicted based on the nonlinear fitting ability of the neural network, thereby overcoming the technical problems that the terrain of the target area is complex, difficult to predict, and the final waypoint height result is inaccurate. Thereby improving the accuracy of determining the waypoint height.

[0019] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present application, but do not constitute a limitation on the embodiments of the present application. In the accompanying drawings:

[0021] Figure 1 A schematic diagram of a process for determining a waypoint altitude according to an embodiment of the present application is shown;

[0022] Figure 2 A schematic diagram of the structure of a latitude and longitude grid according to an embodiment of the present application is schematically shown;

[0023] Figure 3 A schematic diagram of the structure of a device for determining a waypoint altitude according to an embodiment of the present application is shown schematically. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0025] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back...), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0026] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0027] The altitude mentioned in the embodiments of the present application refers to the height of the target position from the mean sea level; the ellipsoid height refers to the height of the target position from the reference ellipsoid surface of the earth; and the elevation difference refers to the height difference between the altitude and the ellipsoid height.

[0028] Figure 1 The flowchart of a method for determining the height of a waypoint according to an embodiment of the present application is schematically shown. The method for determining the height of a waypoint provided in an embodiment of the present application is used to determine the height of a waypoint of an aircraft such as a drone. Figure 1 As shown, an embodiment of the present application provides a method for determining a waypoint altitude, which may include the following steps:

[0029] S120, obtaining a gridded regional elevation difference data set of a target area, a target longitude and latitude of a target waypoint in the target area, and a target altitude; wherein the gridded regional elevation difference data set includes: cell vertex data corresponding to a plurality of cells, the cell vertex data including vertex longitude and latitude data of each cell and actual vertex elevation difference data;

[0030] S140, determining a target cell of the target waypoint among the multiple cells according to the target longitude and latitude and the gridded regional elevation difference data set;

[0031] S160, based on the pre-trained neural network elevation difference prediction model, according to the target longitude and latitude and the cell vertex data corresponding to the target cell, obtain the predicted elevation difference of the target waypoint;

[0032] S180: Determine the ellipsoid height of the target waypoint according to the predicted elevation difference and the target altitude to obtain the waypoint height of the target waypoint.

[0033] The method for determining the altitude of a waypoint provided by the embodiment of the present application is to obtain the target longitude and latitude of the target waypoint and the target altitude. According to the target longitude and latitude, in the gridded area elevation difference data set, the vertex longitude and latitude data of the target cell where the target waypoint is located and the vertex actual elevation difference data of the multiple cell vertices of the target cell are determined. The target longitude and latitude, the vertex longitude and latitude data of the multiple cell vertices and the actual elevation difference data of the vertex are input into the neural network elevation difference prediction model, and the model outputs the predicted elevation difference of the target waypoint accordingly. Subsequently, the ellipsoid height of the target waypoint can be determined according to the predicted elevation difference and the target altitude. Since the terrain of the actual target area is complex and does not change linearly, the inventor uses the neural network elevation difference prediction model to achieve better elevation difference prediction results. Based on the nonlinear fitting ability of the neural network, the elevation difference of any waypoint in the target area is predicted, thereby overcoming the technical problems that the terrain of the target area is complex, difficult to predict, and the final waypoint altitude result is inaccurate. Thereby improving the accuracy of determining the altitude of the waypoint.

[0034] It can be understood that the gridded area is a map obtained by dividing the target area into a grid, and the gridded area includes a plurality of cells defined by cell vertices, each cell vertex having a vertex longitude and latitude and an actual elevation difference. The target area can be, for example, a specific area selected artificially, or the entire earth.

[0035] Specifically, for each cell vertex, its vertex longitude and latitude and actual elevation difference data are aggregated into a field of cell vertex data, and the cell vertex data of the gridded area are aggregated into a gridded area elevation difference dataset.

[0036] In some embodiments of the present application, step S140 includes: determining multiple target cell vertex data with the smallest difference from the target longitude and latitude based on the target longitude and latitude and the gridded area elevation difference dataset; and determining the target cell based on the multiple target cell vertex data.

[0037] Specifically, after determining the target longitude and latitude, the cell vertex data closest to the target longitude can be determined according to the target longitude, and the cell vertex data closest to the target latitude can be determined according to the target latitude. And according to the above determined cell vertex data, the shape of the cell and the number of cell vertices that should be, the target cell where the target waypoint is located is determined.

[0038] In some embodiments of the present application, before obtaining the gridded regional elevation difference dataset in step S120, the method further includes:

[0039] Gridding the target area to generate a latitude and longitude grid of the target area; wherein the latitude and longitude grid includes a plurality of cells, each cell having a plurality of cell vertices containing latitude and longitude data;

[0040] According to the latitude and longitude data of each cell vertex, the actual elevation difference data of each cell vertex is obtained;

[0041] A gridded regional elevation difference dataset is generated based on the latitude and longitude data and the actual elevation difference data of each cell vertex.

[0042] It can be understood that the latitude and longitude grid covers the target area and is used to divide the target area into multiple sub-areas through multiple cells. The range of each sub-area is defined by the cell vertex of its cell, and the position of the cell vertex in the target area is represented by the longitude and latitude. The size and properties of multiple cells can be the same or different.

[0043] In some embodiments of the present application, cells in the latitude and longitude grid may be tiled regular polygon cells.

[0044] It can be understood that the tiled regular polygonal cells include: squares, regular triangles and regular hexagons. The above tiled regular polygonal cells can be combined with each other, for example, a longitude and latitude grid can be formed by combining square cells, regular triangle cells and regular hexagonal cells; or a longitude and latitude grid can be formed by combining square cells and regular triangle cells, etc.

[0045] In order to train the neural network elevation difference prediction model to improve the accuracy of the neural network elevation difference prediction model, or to obtain the neural network elevation difference prediction model according to the preset neural network model training. In some embodiments of the present application, after generating the latitude and longitude grid of the target area, the method for determining the waypoint altitude provided by the embodiment of the present application also includes:

[0046] Select the desired point in each cell;

[0047] Determine the latitude and longitude data of each expected point and the actual elevation difference data of the expected point;

[0048] Generate a training vector for each expected point according to the expected point and the longitude and latitude grid, wherein the training vector includes: longitude and latitude data of the expected point corresponding to the expected point and vertex longitude and latitude data of multiple cell vertices of the cell where the expected point is located and actual elevation difference data;

[0049] The preset neural network model is trained according to the training vector so that the predicted elevation difference output by the preset neural network model approaches the actual elevation difference data of the expected point until the model converges to obtain the neural network elevation difference prediction model.

[0050] It can be understood that the desired point is the coordinate point that replaces the target waypoint during the training period. Taking the cell as a square as an example, a training vector should include the longitude and latitude data of the desired point, the longitude and latitude data of each of the four cell vertices, and the actual elevation difference data. The convergence condition of the model convergence can be, for example, that the number of iterations of the preset neural network model reaches a threshold, the training running time reaches a preset time, or the difference between the output predicted elevation difference and the actual elevation difference data of the desired point is less than a preset difference.

[0051] In some embodiments of the present application, the cells in the latitude and longitude grid are tileable regular polygon cells; the desired point is the center of the cell. Figure 2 As shown, the point with a larger radius is the cell vertex, and the point with a smaller radius is the desired point.

[0052] In other embodiments of the present application, the cells in the latitude and longitude grid can be tiled with cells of various shapes. In this case, the center of the cell is difficult to determine, so it is only necessary to select the desired point at any position in the cell.

[0053] In some embodiments of the present application, there are multiple training vectors, and the multiple training vectors are divided into a training set, a validation set, and a test set; wherein the training set has the largest number of training vectors.

[0054] Specifically, the preset neural network model is trained according to the training set, and after each training cycle, the output result of the preset neural network model is verified according to the verification set, and the parameters of the preset neural network model are adjusted according to the verified output result. After the preset neural network model is trained, the output result of the preset neural network model is evaluated according to the test set. If the difference between the predicted elevation difference and the actual elevation difference in the output result evaluated according to the test set is within the allowable range, the preset neural network model is output as the neural network elevation difference prediction model, otherwise the training continues.

[0055] The following is an exemplary description of the method for determining the waypoint altitude provided in an embodiment of the present application.

[0056] Take the Earth as the target area and grid the Earth:

[0057] Draw a longitude line every 0.04° and a latitude line every 0.02° to divide the earth into a large number of squares (approximately 4km×4km in shape). This operation will generate about 3.72 million squares in China, and record the longitude and latitude of the vertices of these squares in two columns of the csv table.

[0058] Then get the elevation difference: go to the elevation calculator website of the United States Geological Survey, import the CSV table, and automatically generate a column of elevation difference data, which represents the difference between the ellipsoid height and the altitude at the corresponding longitude and latitude.

[0059] Then the training data is preprocessed:

[0060] During the earth gridding stage, cell vertices can be generated. If a finer grid size is used, such as 2km×2km, additional desired points can be generated, and the corresponding elevation difference data are obtained for these points.

[0061] The longitude, latitude, and elevation difference of a desired point and the longitude, latitude, and elevation difference of the four surrounding cell vertices are concatenated into a real-valued vector, which is used as the input of the deep neural network model, and the elevation difference of the desired point is used as the desired output of the network.

[0062] In this way, a huge data set is constructed, and then divided into training set, validation set, and test set in the ratio of 60%, 20%, and 20% for training deep neural network models.

[0063] A grid vertex query database as shown in Table 1 is constructed as a gridded regional elevation difference dataset. The longitude and latitude of the cell vertices and the corresponding elevation difference at a 4km×4km grid granularity in the csv file are imported into the MySQL database. Since the combination of longitude and latitude is unique, the location key they are spliced ​​into is also unique. An index is created for the location key to facilitate fast query of records.

[0064]

[0065] Table 1 Mesh vertex query database

[0066] The above real-valued vector is constructed as the input of the deep neural network model, and the network is trained to output the expected value. Problems such as gradient disappearance and gradient explosion often occur during training, so the Leaky ReLU activation function is selected, whose gradient is never 0, making the gradient flow smoother and the training process converge faster. The training process uses the Adam gradient descent algorithm, which uses both momentum and adaptive gradient ideas. Its deviation correction step can alleviate the cold start problem in the early stage of the algorithm.

[0067] After the training is completed, any target waypoint is specified, and the four surrounding cell vertices are queried in the grid vertex database to construct the above-mentioned real-valued vector and input it into the deep neural network model. The deep neural network model will output the predicted elevation difference of the target waypoint.

[0068] By calling the map terrain service interface, you can get the actual altitude of the target waypoint on the map. Then, according to the actual altitude of the target waypoint and the predicted elevation difference, the ellipsoid height of the target waypoint is determined as the waypoint height.

[0069] The embodiment of the present application also provides a device for determining a waypoint altitude, comprising:

[0070] The spatial coordinate acquisition module 320 is used to acquire a gridded regional elevation difference data set of a target area, a target longitude and latitude of a target waypoint in the target area, and a target altitude; wherein the gridded regional elevation difference data set includes: cell vertex data corresponding to a plurality of cells, the cell vertex data includes vertex longitude and latitude data of each cell and actual elevation difference data of the vertex;

[0071] The cell determination module 340 is used to determine the target cell of the target waypoint among the multiple cells according to the target longitude and latitude and the gridded regional elevation difference data set;

[0072] The elevation difference prediction module 360 ​​is used to obtain the predicted elevation difference of the target waypoint based on the target longitude and latitude and the cell vertex data corresponding to the target cell based on the pre-trained neural network elevation difference prediction model;

[0073] The waypoint altitude determination module 380 is used to determine the ellipsoid altitude of the target waypoint according to the predicted elevation difference and the target altitude to obtain the waypoint altitude of the target waypoint.

[0074] In some embodiments of the present application, the spatial coordinate acquisition module 320 can also be used to: grid the target area to generate a latitude and longitude grid of the target area; wherein the latitude and longitude grid includes multiple cells, each cell has multiple cell vertices containing longitude and latitude data; based on the longitude and latitude data of each cell vertex, obtain the actual elevation difference data of each cell vertex; based on the longitude and latitude data of each cell vertex and the actual elevation difference data, generate a gridded area elevation difference data set.

[0075] In some embodiments of the present application, the spatial coordinate acquisition module 320 can also be used to: select an expected point in each cell; determine the expected point longitude and latitude data and the actual elevation difference data of each expected point; generate a training vector for each expected point based on the expected point and the longitude and latitude grid, wherein the training vector includes: the expected point longitude and latitude data corresponding to the expected point, and the vertex longitude and latitude data and actual elevation difference data of multiple cell vertices of the cell where the expected point is located; train a preset neural network model based on the training vector so that the predicted elevation difference output by the preset neural network model approaches the actual elevation difference data of the expected point until the model converges to obtain a neural network elevation difference prediction model.

[0076] In some embodiments of the present application, in the spatial coordinate acquisition module 320, the cells in the latitude and longitude grid are tileable regular polygonal cells.

[0077] In some embodiments of the present application, in the spatial coordinate acquisition module 320, the cells in the latitude and longitude grid are tileable regular polygonal cells; the desired point is the center of the cell.

[0078] In some embodiments of the present application, the cell determination module 340 is specifically used to determine multiple target cell vertex data with the smallest difference from the target longitude and latitude based on the target longitude and latitude and the gridded area elevation difference data set; and determine the target cell based on the multiple target cell vertex data.

[0079] An embodiment of the present application also provides a waypoint altitude determination device, comprising: a memory configured to store instructions; and a processor configured to call the instructions from the memory and implement the method for waypoint altitude determination provided in any of the above embodiments when executing the instructions.

[0080] An embodiment of the present application also provides a machine-readable storage medium having instructions stored thereon, the instructions being used to enable a machine to execute the method for determining a waypoint altitude provided by any one of the embodiments.

[0081] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0082] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0083] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0084] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0085] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0086] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0087] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0088] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0089] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for determining a waypoint altitude, characterized in that: include: Acquire a gridded regional elevation difference data set of a target area, a target longitude and latitude and a target altitude of a target waypoint in the target area; wherein the gridded regional elevation difference data set includes: cell vertex data corresponding to a plurality of cells, the cell vertex data includes vertex longitude and latitude data and vertex actual elevation difference data of each cell; Determine a target cell of the target waypoint among the multiple cells according to the target longitude and latitude and the gridded regional elevation difference data set; Based on the pre-trained neural network elevation difference prediction model, the predicted elevation difference of the target waypoint is obtained according to the target longitude and latitude and the cell vertex data corresponding to the target cell; The ellipsoid height of the target waypoint is determined according to the predicted elevation difference and the target altitude to obtain the waypoint height of the target waypoint.

2. The method according to claim 1, characterized in that Before obtaining the gridded regional elevation difference dataset, the method further includes: Gridding the target area to generate a longitude and latitude grid of the target area; wherein the longitude and latitude grid includes a plurality of cells, each of which has a plurality of cell vertices containing longitude and latitude data; According to the latitude and longitude data of each cell vertex, obtaining actual elevation difference data of each cell vertex; The gridded area elevation difference dataset is generated according to the latitude and longitude data and the actual elevation difference data of each cell vertex.

3. The method according to claim 2, characterized in that After generating the latitude and longitude grid of the target area, the method further includes: selecting a desired point in each of the cells; Determine the desired point latitude and longitude data and the actual elevation difference data of each desired point; Generate a training vector for each of the desired points according to the desired point and the longitude and latitude grid, wherein the training vector includes: the longitude and latitude data of the desired point corresponding to the desired point, and the vertex longitude and latitude data of a plurality of the cell vertices of the cell where the desired point is located and the actual elevation difference data; The preset neural network model is trained according to the training vector so that the predicted elevation difference output by the preset neural network model approaches the actual elevation difference data of the expected point until the model converges to obtain the neural network elevation difference prediction model.

4. The method according to claim 2 or 3, characterized in that: The cells in the latitude and longitude grid are tileable regular polygonal cells.

5. The method according to claim 4, characterized in that The tileable regular polygonal unit cell is a square unit cell.

6. The method according to claim 3, characterized in that The cells in the latitude and longitude grid are tileable regular polygonal cells; the desired point is the center of the cell.

7. The method according to claim 3, characterized in that There are multiple training vectors, and the multiple training vectors are divided into a training set, a validation set and a test set; wherein the training set has the largest number of training vectors.

8. The method according to claim 1, characterized in that The step of determining the cell where the target waypoint is located according to the target longitude and latitude and in the gridded area elevation difference dataset comprises: According to the target longitude and latitude and the gridded area elevation difference data set, a plurality of target cell vertex data with the smallest difference from the target longitude and latitude are determined; The target cell is determined according to the plurality of target cell vertex data.

9. A waypoint altitude determination device, characterized in that: include: a memory configured to store instructions; as well as A processor is configured to call the instructions from the memory and implement the method for waypoint altitude determination according to any one of claims 1 to 8 when executing the instructions.

10. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores instructions for causing a machine to execute the method for determining a waypoint altitude according to any one of claims 1 to 8.

Images