Large-scale low-altitude conflict detection and early warning method based on multi-scale spatial grid

Abstract

The application discloses a large-scale low-altitude conflict detection and early warning method based on a multi-scale space grid, and comprises the following steps: S1, dividing a target space according to a large-scale grid; S2, judging potential conflicts according to the grid occupation of each large-scale grid based on the position of a low-altitude flying vehicle and the large-scale grid coding, marking potential conflict areas, and excluding non-conflict areas; S3, dividing the large-scale grid marked as a potential conflict area into several medium-scale grids; S4, for each medium-scale grid, predicting a future path and grid occupation based on a time window, screening potential conflict areas, and excluding non-conflict areas; S5, dividing the medium-scale grid marked as a potential conflict area into several small-scale grids; and S6, for each small-scale grid, constructing a fusion risk model by superimposing an environmental risk on a grid occupation state based on a time window and an environmental risk, judging a conflict early warning level according to a preset conflict judgment threshold, and issuing a conflict early warning.

Description

Technical Field

[0001] This invention relates to the field of low-altitude flight safety management, specifically to a method for hierarchical conflict detection and early warning using multi-scale airspace grids for large-scale low-altitude aircraft operation scenarios. Background Technology

[0002] Large-scale low-altitude aircraft conflict detection and early warning mainly revolve around three types of methods: geometric determination methods, probabilistic analysis methods, and data-driven methods. Geometric determination methods primarily include protected area models, velocity barrier methods, and grid methods. Protected area models determine whether conflicts will occur by setting specific protected areas, such as Reich protected areas, cylindrical protected areas, and spherical protected areas. Velocity barrier methods detect conflicts by analyzing the relative velocities between aircraft; if an aircraft falls within a certain sector area, a conflict is considered to have occurred. Grid methods detect conflicts by comparing the overlap of aircraft trajectories within the same time period. Probabilistic analysis methods mainly include approximate analysis methods and Monte Carlo methods. Approximate analysis methods predict flight trajectories based on the aircraft's dynamic model and flight intention, and use mathematical models for conflict detection. Monte Carlo methods calculate the conflict probability by repeatedly simulating the aircraft's heading, speed, and position, statistically analyzing the frequency of aircraft entering protected areas. Data-driven methods utilize machine learning and deep learning algorithms, training models based on large amounts of historical data to predict potential conflicts between aircraft.

[0003] The working principles of existing conflict detection technologies are mainly reflected in two aspects: model building and conflict detection. In terms of model building, geometric determination methods transform the aircraft's trajectory into spatial geometric relationships for judgment by constructing protected area, velocity obstacle, or spatial grid models; probabilistic analysis methods predict future trajectories through dynamic models and uncertainty modeling, representing conflict risk in probabilistic form; and data-driven methods rely on large-scale historical flight data, using machine learning or deep learning models to automatically extract features and perform conflict detection. In terms of conflict detection, geometric determination methods often employ direct calculation and rule comparison, suitable for real-time detection; probabilistic analysis methods often use numerical simulation, trajectory prediction, and statistical calculation, emphasizing the quantification of conflict probability; and data-driven methods use trained prediction models for rapid inference, suitable for large-scale applications in complex environments.

[0004] Existing conflict detection methods suffer from two main shortcomings: First, in high-density, highly dynamic low-altitude operational environments, it is difficult to balance computational complexity and real-time performance. While geometric deterministic methods offer high computational efficiency, they lack flexibility and cannot cope with dynamic environmental changes. Probabilistic analysis methods can quantify conflict risks, but their simulation computational costs are high, making them unsuitable for real-time detection and early warning. Data-driven methods can handle complex environments, but they are overly dependent on training data. Second, existing methods mostly employ single-scale detection strategies and lack adaptive mechanisms for different airspace densities and complex operational environments. In actual low-altitude operational environments, the number and spatial distribution of aircraft vary greatly, making it difficult for existing methods to achieve conflict detection and early warning that balances efficiency and accuracy. Summary of the Invention

[0005] In view of this, the present invention proposes a large-scale low-altitude conflict detection and early warning method based on multi-scale airspace grids to solve the problem that existing conflict detection and early warning methods are difficult to balance efficiency and accuracy in scenarios with large differences in the spatial distribution of low-altitude flight conflicts.

[0006] To address the above problems, one aspect of the present invention proposes the following technical solution:

[0007] A method for large-scale low-altitude conflict detection and early warning based on multi-scale airspace grids includes the following steps: S1, dividing the target airspace into large-scale grids; S2, based on the low-altitude aircraft position and large-scale grid coding, judging potential conflicts according to the grid occupancy of each large-scale grid, marking potential conflict areas, and excluding non-conflict areas; S3, dividing the large-scale grids marked as potential conflict areas into several medium-scale grids; S4, for each medium-scale grid, predicting future paths and grid occupancy based on time windows, filtering potential conflict areas, and excluding non-conflict areas; S5, dividing the medium-scale grids marked as potential conflict areas into several small-scale grids; S6, for each small-scale grid, constructing a fusion risk model that superimposes environmental risk and grid occupancy status based on time windows and environmental risks, judging the conflict early warning level according to a preset conflict judgment threshold, and issuing a conflict early warning.

[0008] Further, in step S2, the following method is used to determine potential conflict areas and non-conflict areas: the large-scale grid code of the low-altitude aircraft is calculated based on its real-time position data, and the number of low-altitude aircraft in each large-scale grid is counted to determine its grid occupancy; when the number of low-altitude aircraft in a large-scale grid is 0, the large-scale grid is determined to be a non-conflict area; when the number of low-altitude aircraft in a large-scale grid is ≥2, the large-scale grid is determined to be a potential conflict area; when the number of low-altitude aircraft in a large-scale grid is 1, it is further determined whether there are other low-altitude aircraft in its neighboring grids: if there are neighboring grids with a number of low-altitude aircraft ≥1, the large-scale grid is determined to be a potential conflict area; if there are no low-altitude aircraft in any of the neighboring grids of the large-scale grid, the large-scale grid is determined to be a non-conflict area.

[0009] Furthermore, in step S2, the calculation of the large-scale grid code of the low-altitude aircraft based on its real-time position data specifically includes: constructing a mapping function to calculate the corresponding grid code based on latitude, longitude, and altitude coordinates; and using the mapping function to obtain the large-scale grid code of each low-altitude aircraft based on its real-time position coordinates.

[0010] Furthermore, step S4 specifically includes: for each low-altitude aircraft, predicting from its planned path... t The future time window Δ starting from the moment T The flight path within the time frame is used to obtain a path grid set; at time... t， Integrate the path grid sets of all low-altitude aircraft and statistically analyze the time window Δ for each mesoscale grid. T The number of low-altitude aircraft and their corresponding aircraft numbers are used to construct a mesoscale grid airspace occupancy map for the given time window. For each mesoscale grid, the number of low-altitude aircraft within the grid is determined based on its airspace occupancy map. When the number of low-altitude aircraft in a mesoscale grid is 0, the grid is considered a conflict-free area. When the number of low-altitude aircraft in a mesoscale grid is ≥2, the grid is considered a potential conflict area. When the number of low-altitude aircraft in a mesoscale grid is 1, it is further determined whether there are other low-altitude aircraft in its neighboring grids: if there are neighboring grids with ≥1 low-altitude aircraft, the grid is considered a potential conflict area; if there are no low-altitude aircraft in any of the neighboring grids of the grid, the grid is considered a conflict-free area.

[0011] Furthermore, in step S4, for the first... i A low-altitude aircraft, which operates within the time window Δ T The internal path grid set is: ;

[0012] in, Indicates the first i Low-altitude aircraft at all times t The encoding of the mesoscale grid in which it is located; Indicates the first i Low-altitude aircraft t The start time is Δ. T Within the time window, the set of grids covered by its flight path; Δ t The sampling time interval;

[0013] For the k Medium-scale grid , its in t The start time is Δ. T Map of airspace occupancy status within the time window for: ;

[0014] in, express t The start time is Δ. T Within the time window k The number of aircraft in a medium-scale grid.

[0015] Furthermore, step S4 also includes: issuing a Level 3 conflict warning for mesoscale grids marked as potential conflict areas.

[0016] Further, step S6 specifically includes: constructing an environmental risk model based on the exponential decay relationship between flight risk and obstacle distance on a small-scale grid, and constructing an environmental risk map based on this environmental risk model; at time... t The path grid sets of all low-altitude aircraft are integrated, and the statistical analysis of each small-scale grid is performed. t The start time is Δ. T The number of low-altitude aircraft occupying the airspace within a time window and their corresponding numbers are used to construct a small-scale grid airspace occupancy status map for that time window. The environmental risk map is then merged with the small-scale grid airspace occupancy status map to construct the fused risk model, and the comprehensive risk value for each small-scale grid is calculated. An environmental risk threshold and a conflict judgment threshold are set, where the conflict judgment threshold is equal to the environmental risk threshold plus 1. The conflict warning level is determined based on the relationship between the comprehensive risk value and the conflict judgment threshold, and a corresponding conflict warning is issued.

[0017] Furthermore, in step S6, let the small-scale grid set occupied by the obstacle be... Then any small-scale grid The environmental risk model is defined as follows:

[0018] ;

[0019] in, Indicates the first k Flight risk probability values ​​for a small-scale grid. Indicates the first k Each small-scale grid represents an obstacle; e It is the natural logarithm; This represents the shortest grid distance between the current non-obstacle small-scale grid and the nearest obstacle small-scale grid; attenuation coefficient. , used to control the decay rate;

[0020] The environmental risk map is as follows: ;

[0021] The small-scale grid spatial occupancy status map is as follows: ;in express t The start time is Δ. T Within the time window k The number of aircraft in a small-scale grid;

[0022] The fusion risk model is as follows: .

[0023] Furthermore, in step S6, the environmental risk threshold is set as follows: , Then when At that time, it was determined that a conflict had occurred between the low-altitude aircraft and a static environmental obstacle;

[0024] The conflict determination threshold is set as follows: ,but:

[0025] For any small-scale grid ,like If two or more low-altitude aircraft are located in the same small-scale grid, or if there is one low-altitude aircraft in the small-scale grid and the distance between that low-altitude aircraft and a static environmental obstacle is less than a preset safety range, then the condition is considered met. At this point, if the small-scale grid is determined to meet the criteria for a Level 1 conflict warning, a Level 1 conflict warning will be issued to all low-altitude aircraft within the small-scale grid.

[0026] For any small-scale grid Calculate the sum of the comprehensive risk values ​​of its neighboring grid. ;

[0027] like and This indicates that the grid There is a low-altitude flying vehicle in the grid, and the grid contains... The distance to static environmental obstacles is greater than the preset safety range, but in the grid If more than one low-altitude aircraft exists in a neighboring grid, or if the distance between more than one neighboring grid and a static environmental obstacle is less than a preset safety range, the criteria for a Level 2 conflict warning are met, and the grid is deemed to be in conflict. A Level 2 conflict has been identified, and a Level 2 conflict warning has been issued.

[0028] The present invention also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can implement the steps of the aforementioned method.

[0029] The beneficial effects of this invention's technical solution are reflected in the following: The large-scale low-altitude conflict detection and early warning method proposed in this invention is based on a multi-scale airspace grid layered detection mechanism. Through a multi-scale grid layered strategy of "rapid filtering with large grids—refined screening with medium grids—precise detection with small grids," it first utilizes the low computational cost of large-scale grids to quickly eliminate most safe airspace, concentrating limited computational resources on a few potential conflict areas for refined analysis, thus achieving efficient localization of potential conflict areas. This "focusing" mechanism fundamentally solves the efficiency bottleneck problem caused by global high-precision calculations, thereby achieving a balance between efficiency and accuracy overall. Simultaneously, by employing a multi-source risk fusion model that combines environmental obstacles and aircraft spatial density risks, it not only considers collision risks between aircraft but also identifies potential conflicts between aircraft and environmental obstacles, achieving more comprehensive safety early warning and making it suitable for complex and ever-changing low-altitude flight scenarios. Attached Figure Description

[0030] Figure 1 This is a flowchart of a large-scale low-altitude conflict detection and early warning method based on multi-scale airspace grids, according to an embodiment of the present invention.

[0031] Figure 2 This is an exemplary schematic diagram of multi-scale spatial grid partitioning according to an embodiment of the present invention.

[0032] Figure 3 This is an exemplary schematic diagram of conflict warning classification according to an embodiment of the present invention. Detailed Implementation

[0033] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. The embodiments provided are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0034] This invention aims to address the uneven spatial distribution of large-scale low-altitude flight conflicts by providing a hierarchical conflict detection and early warning mechanism with three different grid scales. Through large, medium, and small-scale grids, a hierarchical progressive conflict detection process of "rapid filtering → refined screening → accurate detection" is implemented to achieve efficient, accurate detection and graded early warning of large-scale low-altitude aircraft conflicts.

[0035] Please refer to Figure 1 The above is a flowchart of a large-scale low-altitude conflict detection and early warning method based on a multi-scale airspace grid, provided in an embodiment of the present invention. Figure 1 As shown, the method includes the following steps:

[0036] S1. Divide the target airspace into large-scale grids. The large-scale grids perform an initial full airspace division of the target airspace to ensure full spatial coverage.

[0037] S2. Based on the location of the low-altitude aircraft and the large-scale grid coding, potential conflicts are determined according to the grid occupancy of each large-scale grid: if a potential conflict exists in a certain large-scale grid, it is marked as a potential conflict area; if there is no potential conflict, the large-scale grid is excluded as a conflict-free area.

[0038] S3. Divide the large-scale grid marked as a potential conflict area into a medium-scale grid. Specifically, divide each large-scale grid with potential conflict into several medium-scale grids.

[0039] S4. For each mesoscale grid, predict future paths and grid occupancy based on time windows, screen potential conflict areas and issue a Level 3 conflict warning, and exclude non-conflict areas.

[0040] S5. Divide the medium-scale grid marked as a potential conflict area into several small-scale grids;

[0041] S6. For each small-scale grid, construct a fusion risk model that combines environmental risk and grid occupancy status based on time window and environmental risk. Determine the conflict warning level according to the conflict judgment threshold and issue a preset conflict warning.

[0042] The implementation process of the above-described method in this embodiment of the invention can be divided into five parts:

[0043] I. Multi-scale spatial grid construction

[0044] In one specific implementation, an octree structure is used to achieve multi-scale grid hierarchical partitioning and dynamic management of the target airspace. A large-scale grid is used for the initial partitioning of the entire airspace to ensure full spatial coverage. The mesoscale grid is obtained from the large-scale grid through two rounds of octet division, meaning each large-scale grid can be further divided into 64 mesoscale grids. The small-scale grid is obtained from the mesoscale grid through two rounds of octet division, meaning each mesoscale grid can be further divided into 64 small-scale grids. See details [link to implementation details]. Figure 2 It should be noted that this division method is merely exemplary, and those skilled in the art can use other division methods to achieve progressive multi-scale airspace division; this invention does not limit this. In some specific embodiments, the size of the large-scale airspace grid can be set to 160m × 160m × 160m. Using the above exemplary division method, the size of the medium-scale grid is 40m × 40m × 40m, and the size of the small-scale grid is 10m × 10m × 10m. Depending on actual needs, the edge length of the large-scale grid can be selected from 100m to 200m. In some preferred embodiments, a dynamic grid can also be used. Specifically, the grid scale does not need to be fixed in advance, but is dynamically adjusted according to the real-time airspace density. For example, when the aircraft density in a certain area exceeds a threshold, a finer grid layer is automatically triggered for detection in that area.

[0045] It should be understood that while it's possible to directly and uniformly divide the entire spatial domain into grids, this would result in significant redundancy. The octree grid is used here primarily to represent the progressive relationship between large, medium, and small-scale grids. If conflicts arise in the large-scale grid, it can be further divided into eight equal parts to obtain a medium-scale grid for the next scale determination; if there are no conflicts, further subdivision is unnecessary. This significantly reduces computational load while ensuring accurate conflict localization.

[0046] Furthermore, the use of two rounds of eight equal divisions is to differentiate the scale of the grid, enabling faster reduction to a smaller grid scale. Those skilled in the art can use other division methods according to actual needs; this invention does not limit the specific method of grid division.

[0047] In theory, besides using the octree partitioning method described in the previous example, another approach could be to use equal latitude and longitude partitioning.

[0048] II. Conflict Early Warning Classification

[0049] Conflict warning levels are established based on the distance at which low-altitude aircraft meet. In this embodiment of the invention, there are three levels of conflict warning. The closer the low-altitude aircraft are at the point of encounter, the higher the warning level. A schematic diagram of the conflict warning levels is shown below. Figure 3 As shown, the specific classification is as follows:

[0050] (1) Level 1 Conflict Warning (Highest Level): When there are two or more low-altitude aircraft in a small-scale grid, or when there is only one low-altitude aircraft in a small-scale grid but the distance between it and the static environmental obstacle is less than the preset safety range, it is considered that the space overlap is serious and there is a risk of direct collision.

[0051] (2) Level 2 Conflict Warning: When the distance between a low-altitude aircraft and a static environmental obstacle is greater than the preset safety range, but there is more than one low-altitude aircraft in the neighboring grid (such as the 26 neighborhood) of the low-altitude aircraft, or the distance between the low-altitude aircraft and the static environmental obstacle is less than the preset safety range, although there is no direct spatial overlap, it is very likely to form a convergence or approach trend in a short period of time, and there is a high risk of conflict.

[0052] (3) Level 3 conflict warning: When there are two or more low-altitude aircraft in a mesoscale grid, or when there is only one low-altitude aircraft but there are other low-altitude aircraft in its neighboring grid, it indicates that there is a certain approach trend.

[0053] III. Large-scale grids eliminate conflict-free areas

[0054] In the first stage of conflict detection and early warning, a large-scale grid is used to conduct a preliminary screening of the overall airspace, and a coarser-grained spatial division is used to achieve rapid filtering of a large area of ​​airspace.

[0055] 3.1) Calculation of airspace occupancy status

[0056] The large-scale grid code of the low-altitude aircraft is calculated based on its real-time position data, and the airspace occupancy status of the large-scale grid is inferred accordingly.

[0057] First, let's assume a low-altitude aircraft at a certain moment. i The position coordinates are The grid code of the large-scale grid in which it is located can be represented as: ;

[0058] Among them, the function This represents the mapping function for calculating the corresponding grid code based on latitude, longitude, and altitude coordinates. For specific coding rules, please refer to GB / T 40087-2021 "Rules for Geospatial Grid Coding". Based on this, define... Indicates low-altitude aircraft i The corresponding large-scale grid coding, Indicates low-altitude aircraft i The corresponding mesoscale grid coding, Indicates low-altitude aircraft i The corresponding small-scale grid coding.

[0059] Count the number of low-altitude aircraft within each occupied large-scale grid, and define a certain code as... G k The number of low-altitude aircraft in the airspace grid is ,but It can be calculated using the following formula: ;

[0060] in, As an indicator function, when a low-altitude aircraft i Located in the grid The value is 1 when the condition is met, and 0 otherwise. N This represents the total number of low-altitude aircraft currently participating in conflict detection. Similarly, this applies to a specific large-scale grid. ,use and Calculate the number of low-altitude aircraft in a large-scale grid. For a certain mesoscale grid ,use and Calculate the number of low-altitude aircraft in a mesoscale grid. For a specific small-scale grid ,use and Calculate the number of low-altitude aircraft in a small-scale grid. .

[0061] Based on this, a large-scale grid-based airspace occupancy status map is constructed. Each element represents the number of low-altitude aircraft in the corresponding grid: ;

[0062] 3.2) Exclusion of conflict-free areas

[0063] When grid Number of low-altitude aircraft At that time, that is If no low-altitude aircraft are present in the grid, the grid is considered a conflict-free area and excluded.

[0064] When grid Number of low-altitude aircraft At that time, that is If there are more than two low-altitude aircraft in the grid, then the grid is determined to be a potential conflict zone.

[0065] When grid Number of low-altitude aircraft At that time, that is This indicates that a low-altitude aircraft exists within the grid. Further analysis is needed to determine if other low-altitude aircraft exist within its 26-neighborhood grid. The neighborhood is defined as:

[0066] If a neighborhood grid exists Meet the requirements for the number of low-altitude aircraft That is, neighborhood grid If other low-altitude aircraft are still present in the grid, then the grid is considered to be... There is a risk of conflict; otherwise, if all neighboring areas Then it can be determined that the grid This is a conflict-free area, and the grid area can be excluded. x j , y j , z j Represents a grid 3D encoded coordinates x k , y k , z k Represents a grid The three-dimensional encoded coordinates.

[0067] IV. Screening Potential Conflict Areas Using Mesoscale Grids

[0068] For the large-scale grid that has been marked as a potential conflict area in the previous calculation, it is further divided into medium-scale grids. A time window is introduced, that is, from the current moment, path prediction and grid occupancy calculation are performed within a certain time range (a few seconds to tens of seconds) in the future to dynamically construct a map of future airspace occupancy status.

[0069] 4.1) Spatial Occupation Calculation Based on Time Window

[0070] To achieve accurate detection and trend analysis of flight path conflicts for low-altitude aircraft, a three-dimensional mesh airspace occupancy model based on a time window is constructed. The size of the time window is set to Δ. T At any time t For each aircraft numbered i Low-altitude aircraft, which are within the time window Δ T The set of path grids within the range is defined as:

[0071] ;

[0072] in, Indicates the first i Low-altitude aircraft at all times t The encoding of the mesoscale grid in which it is located; Indicates the first i Low-altitude aircraft t The start time is Δ.T Within the time window, the set of grids covered by its flight path; Δ t This represents the sampling time interval.

[0073] It should be understood that the flight paths of low-altitude aircraft, such as drones, are planned in advance. Therefore, based on the aircraft's position and speed, the flight path can be calculated according to the pre-planned route. t A future period of time starting from moment Δ T By determining the location of the drone within the given time frame, we can obtain the grid over which the drone flew during that period, thus generating a set of path grids.

[0074] At any moment t Set the path grid of all low-altitude aircraft By integrating the data, the number of low-altitude aircraft occupying each mesoscale grid within the given time window and their corresponding numbers are statistically analyzed, thereby constructing a mesoscale grid occupancy status map for that time period. The definition is as follows: ;

[0075] in, This means that the grid is a mesoscale grid. exist t The start time is Δ. T A map showing the airspace occupancy status within a given time window.

[0076] 4.2) Screening of potential conflict areas based on time windows

[0077] For a time window Δ T at any time t Perform the following judgment:

[0078] When grid Number of low-altitude aircraft At that time, that is If it is determined that no conflict will occur within the grid, the grid region can be excluded.

[0079] When grid Number of low-altitude aircraft At that time, that is If there are more than two low-altitude aircraft in the grid, the grid is determined to be a potential conflict area and subsequent small-scale grid conflict detection is required.

[0080] When grid Number of low-altitude aircraft At that time, that is This indicates that a low-altitude aircraft exists within the grid. Further analysis is needed to determine if other low-altitude aircraft exist in the mesoscale grids within its 26-dimensional neighborhood. If such neighboring grids exist... Meet the number of aircraft That is, neighborhood grid If other low-altitude aircraft are still present in the grid, then the grid is considered to be... There is a risk of conflict, requiring subsequent small-scale mesh conflict detection; otherwise, if the neighboring mesh... Number of aircraft inside Then determine the grid. This is a conflict-free area, and the grid area can be excluded.

[0081] 4.3) Level 3 Conflict Warning Issued

[0082] Mesoscale grid Number of low-altitude aircraft ,or Furthermore, if other low-altitude aircraft are present in the vicinity, the conditions for a Level 3 early warning are met. In this case, it is necessary to monitor the grid. A Level 3 conflict warning has been issued for all low-altitude aircraft in the region.

[0083] V. Fine-grained conflict detection using small-scale meshes

[0084] Based on the screening of potential conflict areas using a mesoscale grid, the grid is further refined at a small-scale stage to achieve higher accuracy in conflict detection. In addition to airspace occupancy status based on time windows, an environmental risk model is introduced to integrate and construct a multi-source risk perception framework, enabling comprehensive, dynamic, and high-resolution identification of potential conflicts involving low-altitude aircraft in complex low-altitude airspace.

[0085] 5.1) Environmental Risk Model

[0086] To characterize the risks posed to flight safety by static obstacles (such as tall buildings, towers, and no-fly zones), an environmental risk model based on distance attenuation is constructed on a small-scale grid. Let the set of grids occupied by the obstacles be denoted as . Any small-scale grid The environmental risk model is as follows:

[0087] ;

[0088] in, Represents a grid The probability value of flight risk. Represents a grid It is an obstacle; e It is the natural logarithm; This represents the shortest grid distance between the current non-obstacle small-scale grid and the nearest obstacle small-scale grid; attenuation coefficient. This is used to control the decay rate of the function, ensuring that the flight risk changes significantly with distance at shorter distances and decreases with distance at longer distances. This definition reflects the exponential decay relationship between flight risk and obstacle distance; the closer to the obstacle, the higher the risk, while the risk approaches zero at greater distances.

[0089] Based on the above flight risk model, an environmental risk map can be constructed. , is represented as: ;

[0090] In the conflict detection phase of small-scale grids, the environmental risk map can be overlaid with the airspace occupancy status map, thereby enabling the assessment of conflicts between low-altitude aircraft and conflicts with the static environment, achieving more accurate conflict early warning.

[0091] 5.2) Small-scale grid spatial occupancy status map based on time window

[0092] At any moment t Set the path grid of all low-altitude aircraft The data is integrated, and the number of low-altitude aircraft occupying each small-scale grid within the time window and their corresponding numbers are counted to construct a small-scale grid for that time period. Airspace occupancy status map The definition is as follows: ;

[0093] in, express t The start time is Δ. T Within the time window k The number of aircraft in a small-scale grid.

[0094] 5.3) Conflict Detection Based on Integrated Risk Models

[0095] To achieve synergistic analysis of environmental and operational density risks, a unified model is used to model the flight risks posed by static obstacles and the spatial distribution of low-altitude aircraft. This enables the overlay perception of multi-source risks and integrates the environmental risk map. Map of airspace occupancy status Perform fusion and calculate the comprehensive risk value for each small-scale grid:

[0096] ;

[0097] Set the environmental risk threshold as Then when At that time, it is determined that a conflict will occur between the low-altitude aircraft and static environmental obstacles. For example, when the safety range is set to 3 grids, the threshold will be set to... If the environmental risk of a grid exceeds this threshold, it means that the distance between the grid and the static environmental obstacle is less than 3 grids, and a conflict is determined to occur.

[0098] 5.4) Conflict early warning issuance

[0099] Based on the definition of conflict early warning, a conflict judgment threshold is set. Based on the classification of Level 1 and Level 2 conflict warnings, different levels of conflict are determined and warnings are issued. It should be noted that the conflict judgment threshold... This refers to the situation where there is one aircraft in the grid, and the environmental risk in that grid reaches the environmental risk threshold. When there are two or more aircraft in the grid, the comprehensive risk value is ≥2, and will definitely be greater than [a certain threshold]. ε Therefore, when the overall risk value of the grid exceeds the threshold... ε When this happens, it will be classified as a Level 1 conflict.

[0100] (1) Level 1 Conflict Warning Issued

[0101] For any grid ,like If so, it is considered that two or more low-altitude aircraft are located in the same small-scale grid, or that the grid is... There is a low-altitude aircraft and the distance between the low-altitude aircraft and a static environmental obstacle is less than a preset safety range (i.e., the static environmental risk exceeds the threshold). ,Right now At this point, the distance between low-altitude aircraft or between low-altitude aircraft and environmental obstacles is small, reaching the standard for a Level 1 conflict warning, and the grid is determined to be in a state of emergency. There is a Level 1 conflict.

[0102] If grid If a Level 1 conflict is detected, then the conflict is sent to the grid. All low-altitude aircraft issued a Level 1 conflict warning.

[0103] (2) Level II Conflict Warning Issuance

[0104] For any grid The sum of the comprehensive risk values ​​of its 26 neighborhoods The calculation formula is as follows:

[0105] ;

[0106] like and This indicates that the grid There is a low-altitude flying vehicle in the grid, and the grid contains... The distance to static environmental obstacles is greater than the preset safety range, but in the grid If more than one low-altitude aircraft exists in the neighboring grid, or the distance between the aircraft and static environmental obstacles is less than a preset safe range, the criteria for a Level 2 conflict warning are met, and the affected grid is deemed to be in a conflict zone. A Level 2 conflict has been identified, and a Level 2 conflict warning has been issued.

[0107] In a preferred embodiment, the conflict warning can be sent to the corresponding low-altitude aircraft via a low-latency communication link, for example.

[0108] The following is a specific calculation example to illustrate the implementation of the present invention:

[0109] Taking a large-scale grid of 160m×160m×160m as an example, assume there are three low-altitude aircraft A, B, and C in the airspace. The positions of aircraft A, B, and C are (113.9887571°E, 22.5941167°N, 110), B, and C respectively.

[0110] The large-scale grid 3D codes corresponding to the three aircraft are (214076, 49083, 0), (214076, 49083, 0), and (214074, 49084, 1), respectively. Based on the aircraft locations, a large-scale grid airspace occupancy map is constructed, and the relevant grid occupancy details are as follows: , The remaining grid cells have an occupancy of 0 and are directly excluded. It can be seen that the grid (214074, 49084, 1) is a conflict-free grid, meaning that aircraft C will not conflict with other aircraft. The grid (214076, 49083, 0) is a potential conflict grid and requires further conflict detection.

[0111] Spacecraft A and Spacecraft B in the grid (214074, 49084, 1) at time... t The corresponding mesoscale grid 3D encoding sets with a time window of 9s and a sampling interval of 1s are as follows: {(856304, 196333, 2),(856304, 196333, 2), (856304, 196333, 2), (856305, 196333, 2), (856305,196333, 2), (856305, 196333, 2), (856305, 196332, 2), (856305, 196332, 2),(856305, 196332, 2)}, The mesoscale grid occupancy at each time step is calculated as follows: {(856305, 196334, 2), (856305, 196334, 2), (856305, 196334, 2), (856305, 196333, 2), (856305, 196333, 2), (856306, 196333, 2), (856306, 196333, 2), (856306, 196333, 2), (856306, 196333, 2)}. It can be seen that after 1-3 seconds, the mesoscale grid... , Furthermore, the two are neighboring grids, and after 4-6 seconds, the mesoscale grid... After 7-9 seconds, the mesoscale grid... , Furthermore, the fact that the two are neighboring grids indicates that there is a potential conflict at every moment, requiring the issuance of a level-three warning and further fine-tuning at a small scale.

[0112] In this example, with no static obstacles nearby, the small-scale 3D encoding sets for aircraft A and B within a 9-second time window and a 1-second sampling interval are as follows: {(3425217,785335, 11), (3425217, 785336, 11), (3425217, 785337, 11), (3425216, 785338,11), (3425216, 785339, 11), (3425215, 785339, 11), (3425214, 785339, 11),(3425214, 785340, 11), (3425214, 785341, 11)} The following table shows the small-scale occupancy at each time step: {(3425220, 785339, 11),(3425219, 785339, 11),(3425218, 785339, 11),(3425217, 785339, 11),(3425216, 785339, 11),(3425216, 785340, 11),(3425216, 785341, 11),(3425216, 785342, 11),(3425216, 785343, 11)}.

[0113]

[0114] As can be seen from the table above, at the 5th second, the two aircraft are located in the same small-scale grid. Therefore, a Level 1 warning should be issued to aircraft A and B, indicating that the warning occurred 5 seconds later and was located in the small-scale grid (3425216, 785339, 11). At the 4th and 6th seconds, the two aircraft are located in the neighboring grid. Therefore, a Level 2 warning should be issued to aircraft A and B, indicating that the warning occurred 4 seconds and 6 seconds later, respectively.

[0115] In summary, the method of the embodiments of the present invention has the following technical advantages:

[0116] 1. A multi-scale grid hierarchical conflict detection method is adopted, which uses large-scale grid for rapid filtering and elimination of conflict-free areas, medium-scale grid for dynamic screening based on time windows, and small-scale grid for refined conflict detection to carry out large-scale low-altitude flight conflict detection.

[0117] 2. Time window-based path grid mapping and occupancy statistics can simultaneously handle trajectory prediction and conflict risk calculation for a large number of aircraft, adapting to large-scale, dynamically changing low-altitude flight environments.

[0118] The present invention also proposes a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, can implement the steps of the methods described in the foregoing embodiments. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (e.g., CD-ROM, USB flash drive, portable hard drive, etc.) and includes several instructions to cause a computer device (e.g., personal computer, server, or network device, etc.) to execute the steps of the methods in various embodiments of the present invention.

[0119] This invention also provides a large-scale low-altitude conflict detection and early warning system based on multi-scale airspace grids, comprising: a grid division module for dividing the airspace into large-scale, medium-scale, and small-scale grids; a large-scale grid conflict detection module for judging potential conflicts based on the low-altitude aircraft position and large-scale grid coding, according to the grid occupancy of each large-scale grid, marking potential conflict areas, and excluding conflict-free areas; a medium-scale grid conflict detection module for predicting future paths and grid occupancy based on time windows, filtering potential conflict areas, and excluding conflict-free areas; a small-scale grid conflict detection module for constructing a fusion risk model that superimposes environmental risk and grid occupancy status based on time windows and environmental risks, judging the conflict early warning level according to a conflict judgment threshold, and issuing a preset conflict early warning; and an early warning issuance module for: issuing a Level 3 conflict early warning to medium-scale grids marked as potential conflict areas, issuing a Level 2 conflict early warning to small-scale grids judged to meet the Level 2 conflict early warning standard, and issuing a Level 1 conflict early warning to small-scale grids judged to meet the Level 1 conflict early warning standard. The specific conflict early warning classification has been detailed above and will not be repeated here.

[0120] The present invention also provides a low-altitude aircraft equipped with the aforementioned large-scale low-altitude conflict detection and early warning system.

[0121] The above description, in conjunction with specific embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, several equivalent substitutions or obvious modifications can be made without departing from the concept of the present invention, and all such modifications, achieving the same performance or application, should be considered within the scope of protection of the present invention.

Claims

1. A method for large-scale low-altitude conflict detection and early warning based on multi-scale airspace grids, characterized in that, Includes the following steps: S1. Divide the target airspace into a large-scale grid; S2. Based on the low-altitude aircraft position and large-scale grid coding, potential conflicts are determined according to the grid occupancy of each large-scale grid, potential conflict areas are marked, and non-conflict areas are excluded. S3. Divide the large-scale grid marked as a potential conflict area into several medium-scale grids; S4. For each mesoscale grid, predict future paths and grid occupancy based on time windows, filter potential conflict areas, and exclude non-conflict areas. S5. Divide the medium-scale grid marked as a potential conflict area into several small-scale grids; S6. For each small-scale grid, a fusion risk model is constructed based on the time window and environmental risk, which combines environmental risk with grid occupancy status. The conflict warning level is determined according to the preset conflict judgment threshold, and a conflict warning is issued. Step S6 specifically includes: An environmental risk model based on the exponential decay relationship between flight risk and obstacle distance is constructed on a small-scale grid, and an environmental risk map is constructed based on this environmental risk model. At any moment t The path grid sets of all low-altitude aircraft are integrated, and the statistical analysis of each small-scale grid is performed. t The start time is Δ. T The number of low-altitude aircraft occupying the airspace within a time window and their corresponding numbers are used to construct a small-scale grid airspace occupancy status map for that time window. The environmental risk map is fused with the small-scale grid airspace occupancy status map to construct the fused risk model, and the comprehensive risk value of each small-scale grid is calculated. Set an environmental risk threshold and a conflict judgment threshold, wherein the conflict judgment threshold is equal to the environmental risk threshold plus 1; determine the conflict warning level based on the relationship between the comprehensive risk value and the conflict judgment threshold, and issue a corresponding conflict warning. In step S6, let the small-scale grid set occupied by the obstacle be... Then any small-scale grid The environmental risk model is defined as follows: ； in, Indicates the first k Flight risk probability values ​​for a small-scale grid. Indicates the first k Each small-scale grid represents an obstacle; e It is the natural logarithm; This represents the shortest grid distance between the current non-obstacle small-scale grid and the nearest obstacle small-scale grid; attenuation coefficient. , used to control the decay rate; The environmental risk map is as follows: ; The small-scale grid spatial occupancy status map is as follows: ;in express t The start time is Δ. T Within the time window k The number of aircraft in a small-scale grid; The fusion risk model is as follows: .

2. The large-scale low-altitude conflict detection and early warning method as described in claim 1, characterized in that, In step S2, the following method is used to determine potential conflict areas and non-conflict areas: the large-scale grid code of the low-altitude aircraft is calculated based on its real-time position data, and the number of low-altitude aircraft in each large-scale grid is counted to determine its grid occupancy. When the number of low-altitude aircraft in a large-scale grid is 0, the large-scale grid is determined to be a conflict-free region. When the number of low-altitude aircraft in a large-scale grid is ≥2, the large-scale grid is determined to be a potential conflict area; When the number of low-altitude aircraft in a large-scale grid is 1, it is further determined whether there are other low-altitude aircraft in its neighboring grids: if there are neighboring grids with a number of low-altitude aircraft ≥ 1, the large-scale grid is determined to be a potential conflict area; if there are no low-altitude aircraft in any of the neighboring grids of the large-scale grid, the large-scale grid is determined to be a conflict-free area.

3. The large-scale low-altitude conflict detection and early warning method as described in claim 2, characterized in that, Step S2 involves calculating the large-scale grid code of a low-altitude aircraft based on its real-time position data. Specifically, this includes: constructing a mapping function based on latitude, longitude, and altitude coordinates to calculate the corresponding grid code; and using the mapping function to obtain the large-scale grid code of each low-altitude aircraft based on its real-time position coordinates.

4. The large-scale low-altitude conflict detection and early warning method as described in claim 1, characterized in that, Step S4 specifically includes: For each low-altitude aircraft, the predicted path from its planned trajectory... t The future time window Δ starting from the moment T The flight path within the area is used to obtain a path grid set; At any moment t The path grid sets of all low-altitude aircraft are integrated, and the time window Δ is statistically analyzed for each mesoscale grid. T The number of low-altitude aircraft and their corresponding aircraft numbers are used to construct a medium-scale grid airspace occupancy map within this time window. For each mesoscale grid, determine the number of low-altitude aircraft within the grid based on its airspace occupancy status map; When the number of low-altitude aircraft in a mesoscale grid is 0, the mesoscale grid is determined to be a conflict-free region. When the number of low-altitude aircraft in a mesoscale grid is ≥2, the mesoscale grid is determined to be a potential conflict area. When the number of low-altitude aircraft in a mesoscale grid is 1, it is further determined whether there are other low-altitude aircraft in its neighboring grids: if there are neighboring grids with a number of low-altitude aircraft ≥ 1, the mesoscale grid is determined to be a potential conflict area; if there are no low-altitude aircraft in any of the neighboring grids of the mesoscale grid, the mesoscale grid is determined to be a conflict-free area.

5. The large-scale low-altitude conflict detection and early warning method as described in claim 4, characterized in that, In step S4, for the first i A low-altitude aircraft, which operates within the time window Δ T The internal path grid set is: ; in, Indicates the first i Low-altitude aircraft at all times t The encoding of the mesoscale grid in which it is located; Indicates the first i Low-altitude aircraft t The start time is Δ. T Within the time window, the set of grids covered by its flight path; Δ t The sampling time interval; For the k Medium-scale grid , its in t The start time is Δ. T Map of airspace occupancy status within the time window for: ; in, express t The start time is Δ. T Within the time window k The number of aircraft in a medium-scale grid.

6. The large-scale low-altitude conflict detection and early warning method as described in claim 4, characterized in that, Step S4 also includes: issuing a Level 3 conflict warning for mesoscale grids marked as potential conflict areas.

7. The large-scale low-altitude conflict detection and early warning method as described in claim 1, characterized in that, In step S6, the environmental risk threshold is set as follows: , Then when At that time, it was determined that a conflict had occurred between the low-altitude aircraft and a static environmental obstacle; The conflict determination threshold is set as follows: ,but: For any small-scale grid ,like If two or more low-altitude aircraft are located in the same small-scale grid, or if there is one low-altitude aircraft in the small-scale grid and the distance between that low-altitude aircraft and a static environmental obstacle is less than a preset safety range, then the condition is considered met. At this point, if the small-scale grid is determined to meet the criteria for a Level 1 conflict warning, a Level 1 conflict warning will be issued to all low-altitude aircraft within the small-scale grid. For any small-scale grid Calculate the sum of the comprehensive risk values ​​of its neighboring grid. ; like and This indicates that the grid There is a low-altitude flying vehicle in the grid, and the grid contains... The distance to static environmental obstacles is greater than the preset safety range, but in the grid If more than one low-altitude aircraft exists in a neighboring grid, or if the distance between more than one neighboring grid and a static environmental obstacle is less than a preset safety range, the criteria for a Level 2 conflict warning are met, and the grid is deemed to be in conflict. A Level 2 conflict has been identified, and a Level 2 conflict warning has been issued.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program can perform the steps of the method according to any one of claims 1-7.

Images