Dynamic low-altitude airspace intelligent division method, system, equipment and medium

Abstract

The invention discloses a dynamic low-altitude airspace intelligent division method, system and device and a medium, and relates to the technical field of airspace division, and the method comprises the steps: obtaining deterministic region data, including a fixed deterministic region and a temporary deterministic region, and dynamically updating the temporary deterministic region according to a preset rule; based on a targeted data filtering technology, separating point data corresponding to the deterministic region data in the DEM terrain three-dimensional point cloud data in the target region range to obtain a DEM terrain three-dimensional point cloud subset; performing four-dimensional adaptive clustering on the DEM terrain three-dimensional point cloud subset according to a preset total number of airspaces in combination with historical meteorological data of the target area to obtain a plurality of airspace units; and synthesizing the one or more deterministic regions with the plurality of airspace units based on a deterministic region priority strategy to obtain a final airspace division scheme. According to the method, the low-altitude airspace of the complex terrain can be dynamically, efficiently, intelligently and safely divided.

Description

Technical Field

[0001] This application relates to the field of airspace delineation technology, and in particular to a dynamic low-altitude airspace intelligent delineation method, system, device and medium. Background Technology

[0002] The low-altitude economy is developing rapidly, and the scientific planning of low-altitude airspace is its core foundation. Traditional airspace delineation methods either rely on manual experience or use simple grids, making them difficult to adapt to the specific characteristics of certain regions. In particular, some schemes rely on airspace planning experts and GIS tools for airspace delineation, with the decision-making process heavily dependent on subjective human judgment. However, this approach is extremely inefficient and lacks precision. Manually drawn boundaries are difficult to accurately conform to natural obstacles such as mountains and canyons, potentially leading to unreasonable airspace boundaries and failing to fully consider complex and variable meteorological factors, thus posing safety hazards. Furthermore, this method is highly subjective, making rapid and dynamic optimization difficult, and unable to meet the growing demand for low-altitude flights. Summary of the Invention

[0003] The purpose of this application is to provide a dynamic low-altitude airspace intelligent delineation method, system, device and medium, which can dynamically, efficiently, intelligently and safely delineate low-altitude airspace in complex terrain.

[0004] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a dynamic low-altitude airspace intelligent partitioning method, including: Acquire deterministic region data and DEM terrain 3D point cloud data within the target area; wherein, the deterministic region data includes the location and shape data of one or more deterministic regions, the deterministic regions include fixed deterministic regions and temporary deterministic regions, and the temporary deterministic regions are dynamically updated according to preset rules; Based on targeted data filtering technology, the point data corresponding to the deterministic region data within the DEM terrain 3D point cloud data are separated to obtain a subset of the DEM terrain 3D point cloud. Based on the preset total number of airspaces and combined with the historical meteorological data of the target area, four-dimensional adaptive clustering is performed on the three-dimensional point cloud subset of the DEM terrain to obtain multiple airspace units. Based on the deterministic region priority strategy, one or more deterministic regions are combined with multiple airspace units to obtain the final airspace partitioning scheme.

[0005] Secondly, this application provides a dynamic low-altitude airspace intelligent partitioning system, including: The data acquisition module is used to acquire deterministic region data and DEM terrain 3D point cloud data within the target area; wherein, the deterministic region data includes the location and shape data of one or more deterministic regions, the deterministic regions include fixed deterministic regions and temporary deterministic regions, and the temporary deterministic regions are dynamically updated according to preset rules; The data filtering module is used to separate the point data corresponding to the deterministic region data within the DEM terrain 3D point cloud data based on targeted data filtering technology, so as to obtain a subset of the DEM terrain 3D point cloud. The airspace clustering module is used to perform four-dimensional adaptive clustering on the DEM terrain three-dimensional point cloud subset based on the preset total number of airspaces and combined with the historical meteorological data of the target area, so as to obtain multiple airspace units. The final airspace partitioning module is used to synthesize one or more deterministic regions with multiple airspace units based on a deterministic region priority strategy to obtain the final airspace partitioning scheme.

[0006] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a dynamic low-altitude airspace intelligent partitioning method.

[0007] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a dynamic low-altitude airspace intelligent partitioning method.

[0008] According to the specific embodiments provided in this application, the following technical effects are disclosed: First, deterministic regional data and DEM terrain 3D point cloud data are acquired within the target area. The deterministic regions include fixed deterministic regions and temporary deterministic regions. The fixed deterministic regions are known and unchanging, while the temporary deterministic regions are dynamically updated according to preset rules, thus fully considering various needs in real-world scenarios and dynamically adjusting the regions. Second, through targeted data filtering technology, the absolute rigidity and accuracy of the boundaries of all deterministic regions (including fixed and temporary deterministic regions) are ensured. Only for the remaining vast and complex terrain, combined with historical meteorological data of the target area, efficient four-dimensional adaptive clustering is applied to generate "adaptive regions," i.e., multiple spatial units, that highly match the terrain features. Thus, this application achieves the goal of both meeting the mandatory control requirements of key areas and efficiently, intelligently, and safely dividing complex terrain, thereby generating a spatial division scheme that fully meets practical application needs and can be flexibly adjusted.

[0009] In short, this application integrates the application of DEM terrain 3D point cloud data, multi-source coordinate system processing, targeted data filtering, and synthesis of multi-layer results of uncontrolled airspace four-dimensional adaptive clustering with fusion of meteorological risks, forming a fully automated technical solution that simulates and optimizes the realistic delineation logic of "extending the monitoring airspace with the controlled airspace (i.e., the deterministic area) as the center". Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is an application environment diagram of the dynamic low-altitude airspace intelligent partitioning method in one embodiment of this application.

[0012] Figure 2 This is a flowchart illustrating a dynamic low-altitude airspace intelligent partitioning method in one embodiment of this application.

[0013] Figure 3 This is a flowchart illustrating the dynamic low-altitude airspace intelligent partitioning method in another embodiment of this application.

[0014] Figure 4 This is a visual diagram of the final airspace allocation scheme in this application.

[0015] Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0017] This application aims to solve the core technical challenge of how to achieve adaptive partitioning and effective integration of deterministic rules (such as airport control areas) and data-driven approaches in complex environments (such as high altitudes).

[0018] The defined areas mentioned in this application include, but are not limited to, controlled airspace centered on existing aviation infrastructure such as airports, heliports, and landing sites, as well as fixed no-fly zones stipulated by relevant departments. Controlled areas generally refer to airspace providing air traffic control services, flight information services, aviation meteorological services, aeronautical information services, and alerting services for flight activities. Surveillance airspace refers to airspace providing flight information services, aviation meteorological services, aeronautical information services, and alerting services for flight activities. Reporting airspace refers to airspace providing aviation meteorological services and alerting services for flight activities, and providing aeronautical information services according to the needs of general aviation users.

[0019] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] The dynamic low-altitude airspace intelligent partitioning method provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, terminal 101 communicates with server 102 via a network. A data storage system can store the data that server 102 needs to process. The data storage system can be set up independently, integrated into server 102, or placed in the cloud or on another server. Terminal 101 can send deterministic region data and DEM terrain 3D point cloud data within the target area to server 102. After receiving the data, server 102 uses targeted data filtering technology, adaptive clustering, and a deterministic region priority strategy to obtain the final airspace partitioning scheme. Server 102 can then feed back the obtained final airspace partitioning scheme to terminal 101. Furthermore, in some embodiments, the dynamic low-altitude airspace intelligent partitioning method can also be implemented independently by server 102 or terminal 101.

[0021] The terminal 101 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. The server 102 can be implemented using a standalone server or a server cluster consisting of multiple servers, or it can be a cloud server.

[0022] It should be noted that the low-altitude airspace used in this application generally refers to areas with a true altitude of 1000m (inclusive) or less, specifically for the flight activities of low-altitude aircraft (such as drones, helicopters, light aircraft, etc.).

[0023] In one exemplary embodiment, such as Figure 2 and Figure 3 As shown, a dynamic low-altitude airspace intelligent partitioning method is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1Taking server 102 as an example, the process includes steps 201 to 204. This process strictly follows the "rules first, data second" processing logic, ensuring the compliance and scientific nature of the partitioning results.

[0024] Step 201: Obtain deterministic region data and DEM terrain 3D point cloud data within the target area; wherein, the deterministic region data includes the location and shape data of one or more deterministic regions, and the deterministic regions include fixed deterministic regions and temporary deterministic regions.

[0025] Based on existing aviation infrastructure and no-fly zones (which are usually unchanged) stipulated by aviation policies and regulations, a predetermined fixed airspace is established, defining its shape, size, and flight altitude. The boundaries and attributes of this airspace are fixed and unaffected by subsequent adaptive clustering, thus constituting a fixed, deterministic region. The temporary deterministic region is dynamically updated according to predetermined rules, including policy issuance and real-time weather data.

[0026] Location data refers to the latitude and longitude of the region's center, while shape data refers to the region's shape parameters, generally including the rectangle's width, height, and clockwise rotation angle relative to true north. In practical applications, the geometric shape corresponding to deterministic region data can be a rectangle, a circle (by determining the distance from the point to the center), or an arbitrary polygon, which can be adjusted by relevant technical personnel as needed.

[0027] A Digital Elevation Model (DEM) is a data model of the Earth's surface elevation obtained through remote sensing technology, typically used to describe the topographic relief. DEM data is used in low-altitude airspace delineation to represent the three-dimensional spatial characteristics of terrain. In practical applications, this application can acquire a high-precision DEM topographic file, such as ASTER GDEM 30M resolution data.

[0028] In a specific application, the preset rules are real-time meteorological data and / or airspace control data; acquiring deterministic regional data and DEM terrain 3D point cloud data within the target area includes: (1) Obtain the latitude and longitude of the regional center and the regional shape parameters of one or more fixed and deterministic regions.

[0029] (2) When the preset rule is real-time meteorological data, one or more temporary deterministic regions are divided according to the meteorological data of the target area within the preset time period, and the latitude and longitude of the regional center and the regional shape parameters of the temporary deterministic regions are collected.

[0030] During the initial airspace planning phase, by integrating historical and conventional meteorological data, areas with long-term meteorological risks can be identified. These areas need to be avoided or have stricter safety altitudes set when delineating airspace boundaries; this corresponds to the temporary definitive areas in this step. Since weather conditions are variable, these temporary definitive areas are also subject to change.

[0031] In actual flight missions, the low-altitude meteorological data detected in real time by UAVs can be used to temporarily or locally activate, close, or adjust the designated temporary deterministic areas, thereby achieving dynamic updates of the deterministic areas.

[0032] (3) When the preset rule is airspace control data, one or more temporary deterministic regions are divided according to the airspace control data, and the latitude and longitude of the regional center and the regional shape parameters of the temporary deterministic regions are collected.

[0033] In practical applications, there are military no-fly zones and areas that require no-fly zones due to certain reasons. These areas may be lifted over time. To facilitate adjustments, temporary fixed areas are delineated accordingly.

[0034] (4) Load the initial DEM terrain file and read the elevation matrix, then perform data transformation on the elevation matrix to obtain a three-dimensional point cloud dataset under a preset coordinate reference system, and mark it as DEM terrain three-dimensional point cloud data. Among them, the Z dimension represents the altitude, so that the division fully reflects the elevation difference characteristics.

[0035] (5) For any deterministic region, perform coordinate system correction and coordinate system transformation on the regional center latitude and longitude, and combine with the corresponding regional shape parameters to obtain deterministic region data under the preset coordinate reference system. Since the coordinates of the center point of the deterministic region in the directly collected initial deterministic region information are latitude and longitude, and this GCJ-02 latitude and longitude coordinates are encrypted, it is necessary to first decrypt and convert them to the GPS standard coordinate system WGS-84, and then convert them to the projection coordinate system coordinates consistent with the DEM digital elevation model file.

[0036] In short, if the latitude and longitude of the collected deterministic area's center are GCJ-02 coordinates, then it needs to be converted from GCJ-02 to WGS-84 and then projected onto the UTM coordinate system consistent with the DEM to ensure accurate geographical alignment. This process involves coordinate system correction and transformation, and the specific steps are as follows: The data processing steps corresponding to coordinate system correction include: calling the eviltransform library and executing the gcj2wgs_exact function to perform inverse calculation in order to achieve coordinate correction and obtain the latitude and longitude of the region center and the region shape parameters in the standard coordinate system.

[0037] The data processing steps corresponding to the coordinate system transformation include: using the latitude and longitude of the region center and the region shape parameters under the standard coordinate system, calling the pyproj library to create a converter from the standard coordinate system to the preset coordinate reference system and performing coordinate system transformation to obtain deterministic region data under the preset coordinate reference system.

[0038] Step 202 involves separating the point data corresponding to the deterministic region data within the DEM terrain 3D point cloud data based on targeted data filtering technology, thereby obtaining a subset of the DEM terrain 3D point cloud. This step aims to completely separate the rule-driven deterministic region from the data-driven adaptive region at the data level. In a specific application, step 202 includes the following four processing steps.

[0039] (1) Create a Boolean mask array based on the DEM terrain three-dimensional point cloud data; each Boolean mask in the Boolean mask array corresponds to a point data in the DEM terrain three-dimensional point cloud data, that is, it is the same length as the total point cloud; each Boolean mask is initially set to a first value, which can be 0 or 1, or can be represented by True, which means that all points belong to the "adaptive region" to be divided.

[0040] (2) For any deterministic region, based on the location and shape data of the deterministic region, perform an inverse coordinate transformation on the DEM terrain three-dimensional point cloud data so that the geometric shape corresponding to the deterministic region is located at the origin of the coordinate system.

[0041] To achieve efficient judgment, a temporary coordinate system transformation (translation and reverse rotation) is performed on the entire DEM terrain 3D point cloud data. The parameters of this transformation are determined by the center point of the current deterministic region and the rotation angle. This coordinate change is equivalent to "aligning" the geometry corresponding to the deterministic region with the coordinate origin.

[0042] (3) Compare the three-dimensional point cloud data of the DEM terrain after coordinate system transformation with the geometric shape corresponding to the deterministic region, and modify the Boolean mask corresponding to the point data falling within the geometric shape to the second value.

[0043] In the transformed coordinate system, all points falling within the defined region can be accurately and quickly identified by simply comparing the boundaries of the geometric shapes.

[0044] (4) Remove the point data corresponding to the Boolean mask of the second value from the DEM terrain three-dimensional point cloud data to obtain a subset of DEM terrain three-dimensional point cloud.

[0045] In the Boolean mask array, the setting corresponding to True can set the second value of points that have been identified as belonging to the current deterministic region to False. After all deterministic regions have been traversed, the points in the mask whose value is still True constitute a subset of the DEM terrain 3D point cloud that does not belong to any deterministic region and is purely used for adaptive partitioning.

[0046] To visually demonstrate the process of the above-mentioned targeted data filtering technology, this application provides the following numerical calculation process: The known controlled airspace parameters are set as follows: the center point coordinates (cx, cy) are (1000, 2000), the width of the geometric shape corresponding to the deterministic area is 200 meters, the height is 400 meters, and the rotation angle (clockwise from the north reference) is 30 degrees; the terrain point data to be judged is set as follows: point A (assuming it is inside the area): (938.4, 2093.3).

[0047] In the process of inverse coordinate system transformation and precise identification of points within the region, firstly, 30 degrees (clockwise from the North datum) is converted to a reverse rotation angle of -60 degrees (mathematical standard), and the trigonometric function values ​​are calculated: cos(-60°) = 0.5, sin(-60°) = -0.866; then, the following three processes are performed on point A (938.4, 2093.3): 1) Translation processing: (938.4-1000, 2093.3-2000)=(-61.6, 93.3).

[0048] 2) Reverse rotation processing: x'=-61.6 0.5-93.3 (-0.866)=50.0, y'=-61.6 (-0.866)+93.3 0.5 = 100.0.

[0049] 3) Judgment and processing: abs(50)≤100(True), abs(100)≤200(True).

[0050] The final conclusion is that point A is inside.

[0051] Repeat the above steps until the judgment of all point data in the DEM terrain 3D point cloud data is completed.

[0052] Thus far, this application uses an algorithm to separate the predefined deterministic spatial domain from the DEM terrain 3D point cloud data, and the remaining point cloud data is used for subsequent clustering and adaptive partitioning, as detailed in step 203 below.

[0053] Step 203: Based on the preset total number of airspaces and combined with the meteorological data of the target area, perform four-dimensional adaptive clustering on the DEM terrain three-dimensional point cloud subset to obtain multiple airspace units.

[0054] Clustering is a machine learning algorithm that divides a dataset into multiple subsets (clusters). Data points within each cluster have high similarity, while data points between different clusters have low similarity. It is commonly used in data analysis, image processing, and other fields. When performing adaptive clustering on the DEM terrain 3D point cloud subset, the algorithm used is one of the following: Mini-Batch K-Means clustering, DBSCAN algorithm, or hierarchical clustering algorithm.

[0055] Among them, the DBSCAN algorithm can identify clusters of arbitrary shapes and process noise, making it more suitable for exploring spatial regions with unknown shapes; the hierarchical clustering algorithm can obtain partitioning schemes with different granularities; the Mini Batch K Means algorithm can efficiently process large-scale data through batch iteration, while taking into account both clustering accuracy and computation speed, making it particularly suitable for large-scale data computation needs in complex terrains of plateaus.

[0056] Specifically, the Mini Batch K-Means algorithm is an improved version of the K-Means clustering algorithm. It employs a batch update method, which increases the computational speed of clustering and makes it suitable for processing large-scale data. Compared to the traditional K-Means algorithm, Mini Batch K-Means has significant advantages in computational efficiency and memory consumption. The specific processing steps of the Mini Batch K-Means algorithm are as follows: 1. Initialization: Randomly select n points of the preset total number of spatial domains as the initial cluster centers, which are also the spatial domain centers.

[0057] 2. Iterative Loop: 1) Mini-batch, multiple, and non-replacement sampling of sample points; 2) Sample point allocation: For each sample point, calculate the three-dimensional spatial distance (i.e., Euclidean distance) between the sample point and all current cluster centers; then, based on this three-dimensional spatial distance, allocate the sample point to the nearest cluster center; 3) Centroid update: Fine-tune the position of each cluster center based only on the "mini-batch" sample points allocated to each cluster center in the previous step.

[0058] 3. Repeat the above iterative process until all data has been processed.

[0059] The adaptive clustering process described above is performed in three-dimensional space (X,Y,Z) to ensure that the partitioning results fully reflect the topographic relief. The clustering algorithm divides points into multiple clusters, each cluster representing an adaptive spatial unit with similar topographic features.

[0060] In practical applications of this clustering algorithm, this application leverages the framework's excellent scalability, reserving interfaces to elevate the clustering space to higher dimensions. When the preset rules are meteorological data, a value representing the historical wind shear or turbulence frequency of a specific region (i.e., historical meteorological data) is introduced as a fourth dimension, which, together with the three-dimensional coordinates (X,Y,Z), forms a new clustering input (X,Y,Z,...). ), This is an adjustable weighting coefficient used to adjust the degree of influence of meteorological risk factors on the clustering algorithm decision-making. The meteorological imbalance level is determined based on meteorological data within the target area.

[0061] At this point, based on historical meteorological data and the DEM (x,y,z), a four-dimensional clustering algorithm is used to adaptively divide the three-dimensional point cloud subset of the DEM terrain. This ensures that the generated airspace boundary not only conforms to the terrain but also "avoids" areas with high historical meteorological risks. The specific process is as follows: (31) Based on the historical meteorological data of the target area, spatial modeling is performed to obtain a comprehensive spatial meteorological risk field.

[0062] 11) Data source, i.e. the process of acquiring historical meteorological data, includes: acquiring historical meteorological grid reanalysis datasets (ERA5, fifth-generation European Centre for Medium-Range Weather Forecasts climate reanalysis data) covering the target area. These data provide long-term, detailed meteorological parameters, such as wind speed, wind direction, air pressure, turbulence intensity, visibility, etc.

[0063] 12) The spatial modeling process includes: for low-altitude flight safety, based on ERA5 historical reanalysis data (barosphere), quantifying key meteorological risk factors in three-dimensional space, and generating a comprehensive risk field through normalized weighting. Specifically, this includes: Persistent strong wind risk (weight 50%): Calculated based on the historical average wind speed in three-dimensional space using U / V wind components. This indicator is used to identify airspace that is consistently exposed to high wind speeds, directly reflecting the difficulty of maintaining aircraft power and the risk of energy consumption.

[0064] Vertical airflow disturbance risk (weight 30%): Calculates the time series standard deviation of vertical velocity (w component). This indicator quantifies the volatility of vertical airflow and is used to identify areas prone to strong updrafts / downdrafts, serving as a direct basis for judging the intensity of turbulence and disturbances.

[0065] Low visibility and high humidity risk (weight 20%): Based on relative humidity (r) data, the frequency of occurrence of relative humidity >95% over historical periods is statistically analyzed. This indicator serves as a proxy variable for cloud cover, advection fog formation, and potential aircraft icing environments, reflecting the probability of visual flight (VFR) being hindered.

[0066] 13) Generate a risk weight layer w: The above-mentioned multiple risk factors are weighted and fused using a linear weighting method to generate a standardized three-dimensional meteorological risk raster data that is spatially aligned with the DEM data. Each pixel value w (range 0-1) in the raster represents the historical comprehensive meteorological risk index for that geographical location and altitude; the higher the w value, the greater the risk.

[0067] Weighted fusion and normalization: First, the risk factors of the three dimensions are normalized using Min-Max to eliminate differences in dimensions. Then, a linear weighted method is used to fuse and generate a comprehensive risk index. The weight configuration set in the code is: average wind speed (0.5) + vertical disturbance (0.3) + low visibility (0.2) to ensure that dynamic risk dominates.

[0068] Spatial alignment and reprojection: The original meteorological data (EPSG:4326) is reprojected and resampled using bilinear interpolation to make its horizontal coordinate system (CRS), resolution, and geographic boundaries completely consistent with the target DEM topographic data, achieving pixel-level alignment between meteorological and topographic data.

[0069] Vertical coordinate transformation: Based on mean geopotential height data, a mapping relationship between air pressure and altitude is established, transforming the vertical coordinates of meteorological data from "isobaric surfaces (atmospheric pressure unit hPa)" to "mean sea level height (meters)". This ultimately constructs a physically meaningful (x, y, z, w) four-dimensional risk dataset, where w... [0,1].

[0070] (32) Combining the comprehensive spatial meteorological risk field and the DEM terrain 3D point cloud subset, a four-dimensional feature dataset is constructed. Specifically, for each terrain point (x, y, z) in the DEM terrain 3D point cloud subset, its corresponding risk weight w is obtained by spatial interpolation from the meteorological risk raster. Thus, the original 3D point cloud (x, y, z) is upgraded to a four-dimensional feature dataset (x, y, z, ... ).

[0071] (33) Call the mini-batch K-means clustering algorithm to generate multiple spatial units based on the four-dimensional feature dataset.

[0072] Specifically, a scaling factor is introduced. x, y, z are physical coordinates in meters, and their values ​​are much larger than w. To allow meteorological risk to play a role in clustering, a scaling factor must be introduced. . It is a key adjustment parameter that determines the influence of meteorological risk on clustering decisions. The larger the value, the more likely the clustering results will group areas with similar meteorological risks together, even if they are geographically far apart.

[0073] The constructed four-dimensional dataset (x, y, z, ...) The MiniBatchKMeans algorithm considers both spatial proximity and similarity in meteorological risk when calculating "distance".

[0074] This step results in a more intelligent unregulated airspace boundary. For example, a valley with strong crosswinds that is geographically connected to the surrounding area may be classified as a separate airspace unit requiring special attention due to its generally high w value.

[0075] Step 204: Based on a deterministic region priority strategy, one or more deterministic regions are combined with multiple airspace units to obtain the final airspace partitioning scheme. This step is a priority-based labeling process that ensures the absolute priority of deterministic regions. Then, each deterministic region is merged with the airspace units partitioned by the four-dimensional clustering results to generate the final two-dimensional label raster low-altitude visual aeronautical chart.

[0076] In one specific application, step 204 includes: (1) Create an empty final label array; one label in the final label array corresponds to one point data in the DEM terrain three-dimensional point cloud data.

[0077] (2) Assign different first labels to the multiple spatial units obtained after clustering, and fill them into the final label array accordingly.

[0078] (3) For any of the deterministic regions, assign a second label to the point data in the DEM terrain three-dimensional point cloud data that corresponds to the data of the deterministic region, and fill it into the final label array accordingly; wherein, when the point data already has the first label, the second label is used to forcibly cover the first label.

[0079] Specifically, all deterministic regions are traversed again, and all points within each region are found using the same inverse coordinate transformation method. The unique ID (second label) of each deterministic region is then used to forcibly overwrite the existing first labels at these locations. This step ensures that the boundaries of the deterministic regions remain intact and unaffected by the adaptive clustering results.

[0080] (4) Map the filled final label array back to a two-dimensional grid to obtain the final spatial partitioning scheme.

[0081] In practical applications, the method of this application further includes: using the Matplotlib library to overlay the final spatial domain division scheme onto a preset geographic base map (such as a satellite image) for visualization. During visualization, as... Figure 4 As shown, deterministic regions are displayed as precisely rotated rectangles, while adaptive regions are displayed as contour boundaries that fit the terrain. The center points of all regions are clearly marked, visually verifying the rationality of the division.

[0082] In a practical application, after obtaining the final airspace allocation scheme, preset rules (real-time meteorological data, military exercise policy notices, etc.) can be dynamically received to update and adjust the adaptively determined airspace allocation scheme, achieving further updates to the airspace. Specifically, to cope with real-time changes, an operational layer is established on top of the allocation results to achieve airspace activation and management based on real-time meteorological data and dynamic rules. The operational layer includes the following settings: (1) Establish the spatial unit state machine: Establish a status information for each divided airspace unit (whether fixed or adaptively divided), including its number, boundary coordinates, historical meteorological risk level and other static information.

[0083] Define a set of dynamic states, such as "green - normally available", "yellow - proceed with caution" (e.g., requires higher flight qualifications or specific equipment), and "red - temporarily closed".

[0084] Thus, this application achieves the basic visualization of the final airspace partitioning scheme, capable of drawing red rectangles representing deterministic regions and the outlines of adaptive airspace units, and labeling center points and numbers. Furthermore, it can fill the background of these regions with different color saturations based on the average historical risk w of all points within each adaptive airspace unit. For example, regions with higher risk have darker background colors, achieving "risk visualization."

[0085] (2) Construct a dynamic response rule base. The rule base consists of rules hardcoded in the code, which can be manually updated later via files, API calls, etc. The rule base can perform the following processing: Data access: Connect to real-time meteorological data sources, such as weather radar, METAR automatic weather stations, and UAV low-altitude detection data.

[0086] Rule definition (code setting and manual setting): Based on aviation operation standards, set rules to trigger state changes. For example: Rule 1: If the real-time average wind speed in airspace unit #15 is greater than 15 m / s, the status changes from 'green' to 'yellow'.

[0087] Rule 2: If a thunderstorm cell is detected in airspace unit #22, the status immediately changes to 'red'.

[0088] Rule 3: Upon receiving an airspace use instruction for military exercises or emergency rescue, the status of the designated airspace unit can be manually forced to be set to 'red'.

[0089] In a practical management system, the visual interface should be able to change the color (green / yellow / red) of each airspace unit in real time based on dynamic adjustment results, providing intuitive situational awareness. The rules for dynamic adjustment include: policy issuance, real-time weather, etc. For example, when a relevant agency issues an air exercise order involving airspaces #5 and #6, the order is input to the program via file writing or API calls, and the program automatically responds, changing the status of the two airspaces from green (normal) to red (no-fly zone). Another example is that the program itself can be set with a rule (before program execution) that when an airspace is monitored to have an average wind speed exceeding 12.5 m / s over a certain period, the airspace is automatically closed, changing its status from green (normal) to yellow (restricted). Both methods output the current airspace situation map (rendering the corresponding airspace background color on top of the basic low-altitude visual aeronautical chart).

[0090] In short, the temporary deterministic area is dynamically updated based on conditions and rules (temporary military aviation exercises, weather changes, airspace policy regulations, etc.). Since the temporary deterministic area changes, the deterministic area data within the target area also changes, requiring a return to step 201 and repeating steps 201-204 to update the final airspace allocation scheme.

[0091] (3) Enable dynamic updates and publishing: Build a monitoring module to continuously acquire real-time data and evaluate the status of each airspace unit in real time based on the rule base; when the status changes, it is displayed prominently on the visualization interface with different colors.

[0092] The airspace partitioning scheme generated in this application aims to establish a stable airspace benchmark and provide a solid foundation for subsequent "dynamic adjustments." During the actual execution phase of flight missions such as UAVs, a set of dynamic airspace management rules can be established based on real-time low-altitude meteorological data (such as wind fields and precipitation) detected by the UAVs themselves or ground equipment. These rules define the conditions for closing, activating, or adjusting airspace, thereby achieving dynamic airspace management. Based on these rules, the boundary conditions of certain deterministic areas partitioned in this application can be temporarily activated, closed, or adjusted. For example, when the real-time wind speed in a certain area exceeds a safety threshold, the airspace can be temporarily closed, thus achieving refined dynamic airspace management and further ensuring operational safety. Simultaneously, historical conventional meteorological data is added as a fourth dimension to the adaptive clustering algorithm. Based on the consideration of terrain, meteorological data is also included in the airspace partitioning calculation. After the adaptive clustering algorithm is completed, label synthesis is performed to finally generate a basic low-altitude visual aeronautical chart.

[0093] In summary, this application, through an innovative technical solution of "deterministic area dynamic updating - targeted data filtering - adaptive clustering of remaining space - multi-layer result synthesis," simulates and optimizes the realistic delineation logic of "expanding surveillance airspace centered on controlled airspace." This application first ensures the absolute rigidity and accuracy of the boundaries of the "deterministic area" (controlled airspace) for key facilities such as airports and landing sites. Then, it applies efficient automatic clustering to the remaining vast and complex terrain to generate "adaptive areas" (surveillance / reporting airspace) that highly match the terrain features. Thus, this application achieves the goal of both meeting the mandatory control requirements of key areas and efficiently, intelligently, and safely delineating complex plateau terrain, generating an airspace delineation scheme that provides a framework for integrating dynamic meteorological factors and fully meets practical application needs.

[0094] This application presents an intelligent airspace partitioning method that effectively integrates deterministic rules with data-driven adaptive partitioning through a comprehensive step-by-step approach. Its core lies in a partitioning principle of "filtering first, then clustering, and step-by-step synthesis." This application employs a targeted data filtering technique based on inverse coordinate transformation. By temporarily translating and rotating the coordinates of the global point cloud for each deterministic region, it quickly and accurately identifies and marks all data points located within that region in the transformed coordinate system. This application introduces a two-layer meteorological model: a static meteorological risk model is used to optimize the initial airspace boundary delineation and participate in clustering; a dynamic meteorological response model is used to achieve real-time, refined airspace management and dynamic adjustment of temporary deterministic regions. This application uses a priority-based multi-layer label synthesis method. First, clustering labels are assigned to adaptive regions, and then the labels of deterministic regions are used to forcibly cover the corresponding locations, thus ensuring the absolute priority and boundary integrity of the "controlled airspace" definition at the software level.

[0095] Based on the same inventive concept, this application also provides a system for implementing the methods described above. The solution provided by this system is similar to the solution described in the methods above; therefore, specific limitations in one or more system embodiments provided below can be found in the limitations of the methods described above, and will not be repeated here.

[0096] In one exemplary embodiment, a dynamic low-altitude airspace intelligent partitioning system is provided, comprising: The data acquisition module is used to acquire deterministic region data and DEM terrain 3D point cloud data within the target area; wherein, the deterministic region data includes the location and shape data of one or more deterministic regions, the deterministic regions include fixed deterministic regions and temporary deterministic regions, and the temporary deterministic regions are dynamically updated according to preset rules.

[0097] The data filtering module is used to separate the point data corresponding to the deterministic region data within the DEM terrain 3D point cloud data based on targeted data filtering technology, so as to obtain a subset of the DEM terrain 3D point cloud.

[0098] The airspace clustering module is used to perform four-dimensional adaptive clustering on the DEM terrain three-dimensional point cloud subset based on the preset total number of airspaces and combined with the historical meteorological data of the target area, so as to obtain multiple airspace units.

[0099] The final airspace partitioning module is used to synthesize one or more deterministic regions with multiple airspace units based on a deterministic region priority strategy to obtain the final airspace partitioning scheme.

[0100] This application integrates the aforementioned integrated dynamic low-altitude airspace intelligent partitioning system, which incorporates fully automated modules from DEM data loading, multi-source coordinate system processing, targeted data filtering, efficient 3D clustering based on Mini Batch K Means, multi-layer result synthesis to visualization output, forming a complete technical solution.

[0101] Compared with the prior art, this application has the following advantages: (1) Significantly improves the compliance and practicality of the partitioning scheme: This application ensures that the controlled airspace boundaries of key facilities such as airports strictly comply with regulatory requirements by pre-setting areas and mandatory coverage, so that the partitioning results can be directly applied to real airspace management, solving the fundamental problem that the results of pure clustering algorithms are "good-looking but unusable".

[0102] (2) Significantly improved the efficiency and intelligence of delineation in complex plateau terrain: For the vast plateau region with dramatic topographic relief, this application uses the efficient Mini Batch K Means algorithm to replace the inefficient and subjective manual delineation. The algorithm automatically explores terrain features, realizing rapid, objective, and intelligent delineation of the vast monitoring / reporting airspace.

[0103] (3) Enhanced adaptability and safety of airspace division to terrain environment: The automatic clustering part of this application is carried out in three-dimensional space (including elevation), fully considering the influence of multiple factors such as local meteorology, and supporting the introduction of meteorological data as an expandable fourth dimension, realizing the consideration of long-term meteorological risks. The generated boundary can naturally fit the terrain such as ridges and valleys, which is crucial for planning safe low-altitude routes and avoiding ground obstacles in plateau areas. It is more scientific and safer than traditional two-dimensional grid division.

[0104] (4) High flexibility and dynamic adaptability: It should be noted that airspace management needs are dynamic. With this application, managers only need to modify the parameter list of the preset area (such as adding a temporary take-off and landing point, which can correspond to adding a temporary deterministic area) to quickly regenerate a complete and coordinated division scheme; in addition, this application can seamlessly connect to the real-time meteorological system and support the temporary activation, closure or adjustment of airspace according to preset "rules", which has a strong dynamic adaptability and rapid response capability.

[0105] (5) Scientific airspace delineation must consider multiple factors such as available flight path altitude, ground obstacles, infrastructure coverage, and even key meteorological risks, and must be effectively linked with national airspace classifications (such as controlled, monitored, and reported airspace). Therefore, the intelligent delineation method provided in this application, which integrates terrain features and key facility layouts, is of great significance for ensuring the safety of low-altitude flights in high-altitude areas and improving the efficiency of airspace resource utilization.

[0106] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 5As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media to run. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a dynamic low-altitude airspace intelligent partitioning method.

[0107] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0108] In one exemplary embodiment, a computer device is also provided, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the above-described method embodiments.

[0109] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0110] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0111] 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, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0112] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0113] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0114] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0115] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A dynamic low-altitude airspace intelligent partitioning method, characterized in that, The method includes: Acquire deterministic region data and DEM terrain 3D point cloud data within the target area; wherein, the deterministic region data includes the location and shape data of one or more deterministic regions, the deterministic regions include fixed deterministic regions and temporary deterministic regions, and the temporary deterministic regions are dynamically updated according to preset rules; Based on targeted data filtering technology, the point data corresponding to the deterministic region data within the DEM terrain 3D point cloud data are separated to obtain a subset of the DEM terrain 3D point cloud. Based on the preset total number of airspaces and combined with the historical meteorological data of the target area, four-dimensional adaptive clustering is performed on the three-dimensional point cloud subset of the DEM terrain to obtain multiple airspace units. Based on the deterministic region priority strategy, one or more deterministic regions are combined with multiple airspace units to obtain the final airspace partitioning scheme.

2. The dynamic low-altitude airspace intelligent partitioning method according to claim 1, characterized in that, The preset rules are real-time meteorological data and / or airspace control data; Acquire deterministic regional data and DEM terrain 3D point cloud data within the target area, including: Obtain the latitude and longitude coordinates of the regional center and the regional shape parameters of one or more fixed and deterministic regions; When the preset rule is real-time meteorological data, based on the meteorological data of the target area within a preset time period, one or more temporary deterministic areas are divided, and the latitude and longitude of the regional center and the regional shape parameters of the temporary deterministic areas are collected. When the preset rule is airspace control data, one or more temporary deterministic regions are divided according to the airspace control data, and the latitude and longitude of the regional center and the regional shape parameters of the temporary deterministic regions are collected. Acquire the initial DEM terrain file within the target area; The initial DEM terrain file is loaded and the elevation matrix is ​​read. Then, the elevation matrix is ​​converted to obtain a three-dimensional point cloud dataset under a preset coordinate reference system, and it is marked as DEM terrain three-dimensional point cloud data. For any given region, coordinate system correction and transformation are performed on the latitude and longitude of the region center. Combined with the corresponding region shape parameters, the region data under the preset coordinate reference system is obtained.

3. The dynamic low-altitude airspace intelligent partitioning method according to claim 2, characterized in that, The data processing steps corresponding to coordinate system correction include: calling the eviltransform library and executing the gcj2wgs_exact function to perform inverse calculation in order to achieve coordinate correction and obtain the latitude and longitude of the region center and the region shape parameters in the standard coordinate system. The data processing steps corresponding to the coordinate system transformation include: using the latitude and longitude of the region center and the region shape parameters under the standard coordinate system, calling the pyproj library to create a converter from the standard coordinate system to the preset coordinate reference system and performing coordinate system transformation to obtain deterministic region data under the preset coordinate reference system.

4. The dynamic low-altitude airspace intelligent partitioning method according to claim 1, characterized in that, Based on targeted data filtering technology, point data corresponding to the deterministic region data within the DEM terrain 3D point cloud data are separated to obtain a subset of the DEM terrain 3D point cloud, including: A Boolean mask array is created based on the DEM terrain 3D point cloud data; each Boolean mask in the Boolean mask array corresponds to a point data in the DEM terrain 3D point cloud data, and each Boolean mask is initially set to a first value; For any given deterministic region, based on the location and shape data of the deterministic region, the DEM terrain 3D point cloud data is subjected to an inverse coordinate system transformation so that the geometric shape corresponding to the deterministic region is located at the origin of the coordinate system. The three-dimensional point cloud data of the DEM terrain after coordinate system transformation is compared with the geometric shape corresponding to the deterministic region, and the Boolean mask corresponding to the point data falling within the geometric shape is modified to the second value; The point data corresponding to the Boolean mask of the second value are removed from the DEM terrain 3D point cloud data to obtain a subset of the DEM terrain 3D point cloud.

5. The dynamic low-altitude airspace intelligent partitioning method according to claim 1, characterized in that, When performing adaptive clustering on the DEM terrain 3D point cloud subset, the algorithm used is one of the following: mini-batch K-means clustering algorithm, DBSCAN algorithm, or hierarchical clustering algorithm.

6. The dynamic low-altitude airspace intelligent partitioning method according to claim 1, characterized in that, Based on the preset total number of airspaces and combined with historical meteorological data of the target area, four-dimensional adaptive clustering is performed on the DEM terrain three-dimensional point cloud subset to obtain multiple airspace units, including: Based on historical meteorological data of the target area, spatial modeling is performed to obtain a comprehensive spatial meteorological risk field; By combining the comprehensive spatial meteorological risk field and the DEM terrain 3D point cloud subset, a four-dimensional feature dataset is constructed. The mini-batch K-means clustering algorithm is invoked to generate multiple spatial units based on the four-dimensional feature dataset.

7. The dynamic low-altitude airspace intelligent partitioning method according to claim 1, characterized in that, Based on a deterministic region-first strategy, one or more deterministic regions are combined with multiple airspace units to obtain a final airspace allocation scheme, including: Create an empty final label array; each label in the final label array corresponds to a point data in the DEM terrain 3D point cloud data; Different first labels are assigned to the multiple spatial units obtained after clustering, and these labels are then filled into the final label array accordingly. For any of the aforementioned deterministic regions, a second label is assigned to the point data within the DEM terrain 3D point cloud data corresponding to the data of the deterministic region, and the label is filled into the final label array accordingly; wherein, when the point data already has the first label, the second label is used to forcibly cover the first label; The filled final label array is mapped back to a two-dimensional raster to obtain the final spatial partitioning scheme.

8. A dynamic low-altitude airspace intelligent partitioning system, characterized in that, The system includes: The data acquisition module is used to acquire deterministic region data and DEM terrain 3D point cloud data within the target area; wherein, the deterministic region data includes the location and shape data of one or more deterministic regions, the deterministic regions include fixed deterministic regions and temporary deterministic regions, and the temporary deterministic regions are dynamically updated according to preset rules; The data filtering module is used to separate the point data corresponding to the deterministic region data within the DEM terrain 3D point cloud data based on targeted data filtering technology, so as to obtain a subset of the DEM terrain 3D point cloud. The airspace clustering module is used to perform four-dimensional adaptive clustering on the DEM terrain three-dimensional point cloud subset based on the preset total number of airspaces and combined with the historical meteorological data of the target area, so as to obtain multiple airspace units. The final airspace partitioning module is used to synthesize one or more deterministic regions with multiple airspace units based on a deterministic region priority strategy to obtain the final airspace partitioning scheme.

9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the dynamic low-altitude airspace intelligent partitioning method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the dynamic low-altitude airspace intelligent partitioning method as described in any one of claims 1-7.

Images