Low-altitude airspace management method and system based on data analysis

Abstract

The invention discloses a low-altitude airspace management method and system based on data analysis, and relates to the technical field of air traffic management and low-altitude airspace planning, and the method comprises the steps: 1, multi-scale dynamic grid division: adjusting the grid granularity in real time according to the aircraft density and environment complexity, generating a multi-scale grid by adopting an optimized quadtree data structure, and carrying out the multi-scale dynamic grid division; boundary smoothing processing is carried out on boundary areas of the grids with different scales; 2, intelligent airspace layering: dynamically adjusting the thickness of a height layer according to a risk score, and setting a buffer layer to enhance the flight safety; and 3, intelligent semantic annotation: giving dynamic semantic information such as risk levels and traffic priorities to the grid units, embedding airspace control regulations and flight rules, and realizing multi-source data fusion and real-time updating of semantic information. According to the method, the problems of static division limitation, boundary problem, insufficient real-time performance and the like in the prior art are solved, and the airspace resource utilization rate and the flight safety are improved.

Description

Technical Field

[0001] The present invention relates to the field of aviation traffic management and low-altitude airspace planning technology, and in particular to a low-altitude airspace management method and system based on data analysis. Background Art

[0002] With the rapid development of drone swarms and Urban Air Mobility (UAM), the utilization and complexity of low-altitude airspace have increased significantly. Existing airspace planning technologies have the following deficiencies in terms of flexibility, real-time and refined management:

[0003] 1. Limitations of static division: Traditional airspace management technology relies on fixed grid division and high-level stratification, which cannot adapt to the dynamic changes of high-density aircraft and complex environments. This leads to low utilization of airspace resources and cannot meet the real-time changing airspace management needs.

[0004] 2. Prominent boundary problems: Discontinuities are easily generated in the boundary areas between layers and grids, increasing flight safety risks. Existing methods lack smoothing of boundary areas and cannot effectively eliminate transition problems between grids of different scales and altitude layers.

[0005] 3. Lack of real-time performance: The existing system is unable to respond to the complex and changing airspace environment in a timely manner and cannot meet dynamic mission requirements, which affects the efficiency and safety of flight missions.

[0006] Based on this, it is necessary to provide an efficient low-altitude airspace management method based on multi-scale dynamic grid division and intelligent airspace stratification to solve the problems of rough airspace division, poor dynamic adaptability and low flight safety in the existing technology. Summary of the invention

[0007] The purpose of the present invention is to provide a low-altitude airspace management method and system based on data analysis to solve the problems raised in the prior art.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A low-altitude airspace management method based on data analysis, including the following contents:

[0010] Step S1: Multi-scale dynamic grid division:

[0011] By acquiring real-time regional characteristic data, aircraft density data and environmental complexity data are obtained; aircraft density data and environmental complexity data are standardized to obtain aircraft density scores and environmental complexity scores; grid size is calculated based on aircraft density scores and environmental complexity scores to obtain dynamic grid size; multi-scale grids are generated using an optimized quadtree data structure to obtain multi-scale grid data; the boundaries of multi-scale grids are smoothed using the Bezier curve interpolation method to obtain smoothly transitioned grid boundary data; ultimately, the generated multi-scale grid can not only adapt to complex airspace environments, but also lay an efficient data foundation for subsequent hierarchical airspace management and intelligent semantic labeling.

[0012] Step S2: Intelligent airspace stratification:

[0013] The risk score is calculated based on the aircraft density score, environmental complexity score and task priority score to obtain the risk score data; the altitude layer thickness is calculated based on the risk score data to obtain the dynamically adjusted altitude layer thickness; the buffer layer thickness is calculated based on the aircraft speed and the safety time constant to obtain the buffer layer thickness data; the airspace is stratified based on the risk score data and the buffer layer thickness data to obtain an intelligent stratified airspace model; by dynamically calculating the buffer thickness, the transition of the aircraft between adjacent layers is ensured to be safe and reliable. The overall stratification technology significantly improves resource utilization and flight safety, and is particularly suitable for the needs of airspace allocation flexibility and reliability in the multi-task execution environment of drones.

[0014] Step S3: Intelligent semantic annotation:

[0015] According to the dynamic labeling rules, the grid units are labeled with risk levels and traffic priorities to obtain preliminary semantic labeling data; according to the information of airspace control regulations, no-fly zones, restricted-fly zones and recommended flight channels, the grid semantics are embedded to obtain semantic labeling data with embedded regulatory information; the semantic information is updated in real time through the knowledge graph and rule engine to obtain real-time updated semantic labeling data. This intelligent semantic labeling technology not only improves the level of refinement of airspace management, but also significantly improves the execution efficiency of complex flight missions.

[0016] Step S11: Real-time regional characteristic evaluation: obtain aircraft density data and environmental complexity data through the actual regional characteristic data obtained; perform standardization processing on the aircraft density data and environmental complexity data to obtain an aircraft density score and an environmental complexity score;

[0017] Step S12: Dynamic grid size calculation: According to the aircraft density score and the environmental complexity score, the grid size S is calculated using the following formula: 网格 :

[0018]

[0019] Among them, S 基础 is the basic grid size, α and β are adjustment coefficients;

[0020] Step S13: generating a multi-scale grid: generating a multi-scale grid using an optimized quadtree data structure according to the dynamically adjusted grid size data, and obtaining multi-scale grid data;

[0021] Step S14: Boundary smoothing: Based on the multi-scale grid data, a Bezier curve interpolation method is used to perform smooth transition processing on the boundary areas of different grid scales to obtain smooth grid boundary data with continuous and consistent boundaries.

[0022] The real-time regional characteristics assessment further includes the following:

[0023] Step S111: Data collection: collecting aircraft density data and environmental complexity data in real time through multi-source sensors;

[0024] Step S112: Data preprocessing: noise filtering, outlier processing and data cleaning are performed on the collected data to obtain high-quality preprocessed data;

[0025] Step S113: Feature extraction: extract key feature indicators from the preprocessed data, including the aircraft movement speed and the environment change rate, to obtain the aircraft density score and the environment complexity score; wherein, the aircraft density score is calculated by a statistical method based on area division, the airspace is divided into several sub-areas, the number of aircraft appearances per unit time in each sub-area is counted, and then the density value is mapped to a score according to a preset scoring rule; the environment complexity score takes into account the terrain complexity, the number of obstacles and the dynamic obstacle factors, and uses a method based on multi-factor weighted average to calculate the environment change rate, and then according to the preset scoring rule, the environment change rate is mapped to a score.

[0026] The dynamic grid size calculation further includes the following:

[0027] Step S121: Adaptive adjustment mechanism: dynamically adjust the adjustment coefficients α and β according to the real-time regional characteristic evaluation results and system load conditions to obtain optimized adjustment coefficients;

[0028] Step S122: Error correction: performing error detection on the calculated grid size, and using an error correction algorithm to correct the grid size to obtain corrected grid size data;

[0029] Step S123: Multi-level optimization algorithm: combining global optimization and local optimization strategies, a multi-level optimization algorithm is used to optimize the grid division to obtain optimized multi-scale grid data.

[0030] The intelligent airspace layering further includes the following contents:

[0031] Step S21: Risk score calculation: The risk score is calculated by integrating the aircraft density score, the environmental complexity score, and the task priority score according to the weights w1, w2, and w3. The formula is as follows:

[0032] Risk score = (w1×aircraft density score) + (w2×environmental complexity score) + (w3×mission priority score)

[0033] Among them, w1+w2+w3=1, and after calculation through the formula, the risk score data is obtained;

[0034] Step S22: Dynamic adjustment of height layer thickness: Calculate the height layer thickness H according to the following formula: 层厚 :

[0035] H 层厚 =H 基础 ×(1-(γ×risk score))

[0036] Among them, H 基础 is the thickness of the base layer, γ is the risk adjustment coefficient, and the dynamically adjusted height layer thickness data is obtained through calculation;

[0037] Step S23: Buffer layer setting: according to the aircraft speed V 飞行器 and safety time constant T 安全 To calculate the buffer layer thickness H 缓冲 , the formula is as follows:

[0038] H 缓冲 =V 飞行器 ×T 安全 .

[0039] The risk score calculation further includes the following contents:

[0040] Step S211: Weight allocation: According to the management requirements and safety standards of different regions, w1, w2 and w3 are reasonably allocated to obtain weight allocation data;

[0041] Step S212: weighted average method: the weighted average method is used to combine the aircraft density score, environmental complexity and task priority score to obtain a comprehensive risk score;

[0042] Step S213: Dynamic adjustment: Dynamically adjust the weight distribution according to the real-time airspace environment and flight mission requirements, optimize the flexibility and adaptability of the risk score, and obtain the adjusted risk score data.

[0043] The intelligent semantic annotation further includes the following contents:

[0044] Step S31: Dynamic labeling rules: assign risk level and traffic priority information to grid cells, and adjust labeling rules based on real-time data;

[0045] Step S32: semantic rule embedding: embed the information of airspace control regulations, no-fly zones, restricted-fly zones and recommended flight channels into the grid semantics, and update them in real time using the knowledge graph and rule engine.

[0046] The semantic rule embedding further includes the following contents:

[0047] Step S321: Regulation parsing: parsing the latest airspace control regulations and converting them into semantic rules that can be processed by the system to obtain parsed regulation data;

[0048] Step S322: knowledge graph construction: by constructing a knowledge graph based on the parsed regulations and real-time data, establishing the association between airspace areas and aircraft to obtain knowledge graph data;

[0049] Step S323: Rule engine generation: Dynamically generate and update semantic rules based on the knowledge graph and real-time data to ensure the accuracy of semantic information and obtain real-time updated semantic rule data.

[0050] Step S324: semantic verification: verify and correct the embedded semantic information, use a verification algorithm to ensure the correctness and consistency of the semantic annotation, and obtain verified semantic annotation data.

[0051] To achieve the above object, the present invention provides the following technical solutions:

[0052] A low-altitude airspace management system based on data analysis, including the following contents:

[0053] The system includes a data processing and grid generation module, an intelligent airspace stratification module, a semantics assignment and update module, and a control and decision module;

[0054] The data processing and grid generation module is used to execute the multi-scale dynamic grid division step, responsible for the collection and processing of real-time data and the generation and optimization of multi-scale grids, and obtains multi-scale grid data and grid boundary data after boundary smoothing;

[0055] The intelligent airspace stratification module is used to execute the intelligent airspace stratification steps, is responsible for risk score calculation, altitude layer thickness adjustment and buffer layer setting, realizes dynamic stratification management of airspace, and obtains an intelligent stratified airspace model;

[0056] The semantic assignment and update module is used to execute the intelligent semantic annotation step, responsible for semantic information annotation, rule embedding and real-time update of semantic information of grid units, and obtains real-time updated semantic annotation data;

[0057] The control and decision module is used to make airspace management decisions and controls based on multi-scale grid data, intelligently layered airspace models and real-time updated semantic annotation data to ensure the normal operation of aircraft in low-altitude airspace.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] 1. Refined airspace management: Through dynamic grid division technology, the grid granularity is adjusted in real time according to the density of aircraft and the complexity of the environment, and a multi-scale grid model is generated. The grid is refined in areas with dense aircraft to improve navigation accuracy, and the grid is expanded in simple areas to optimize the calculation burden, reduce the risk of conflict, improve the efficiency of flight mission planning, and lay a foundation for accurate data for subsequent management;

[0060] Intelligent airspace stratification technology dynamically adjusts the thickness and distribution of airspace layers based on multi-dimensional scoring, increases the number of layers and reduces the layer thickness in high-risk areas, and performs the opposite operation in low-risk areas, optimizes resource allocation, and buffer layer design ensures safe transition between layers, which meets the requirements of airspace flexibility and reliability for drone multi-task execution;

[0061] Intelligent semantic annotation gives grid units rich semantic information, embeds regulatory content, and uses knowledge graphs and dynamic rule updates to adapt to real-time airspace changes, improve the level of refinement of airspace management, and help efficiently execute complex flight missions.

[0062] 2. Improve flight safety: Multi-scale grid division effectively reduces the risk of conflict between aircraft and between aircraft and the environment. Grid boundary smoothing ensures coherence and consistency, avoiding safety hazards caused by grid defects.

[0063] The risk score in intelligent airspace stratification accurately reflects the airspace risk, and based on this, high-risk areas are stratified and reasonable buffer layers are set. Combined with dynamic adjustments, the probability of accidents such as aircraft collisions is significantly reduced, ensuring the safety and reliability of all aspects of flight.

[0064] 3. Enhanced adaptability and flexibility: Data collection integrates multi-source sensor data, with wide coverage and high accuracy, adapting to complex airspace, and data preprocessing to ensure efficient and reliable input; dynamic grid size calculation flexibly adjusts the granularity according to data to adapt to different airspace scenarios, such as dense activities or low-complexity areas;

[0065] The risk score calculation can dynamically adjust weights and update scores as needed, optimize each link in layers as needed, design and optimize the buffer layer as needed, and update semantic annotations as the airspace changes in real time, fully meeting the needs of dynamic changes in the airspace. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1A method flow diagram of a low-altitude airspace management method based on data analysis of the present invention;

[0067] Figure 2 A schematic diagram of multi-scale dynamic grid division of a low-altitude airspace management method based on data analysis of the present invention;

[0068] Figure 3 A schematic diagram of intelligent airspace layering of a low-altitude airspace management method based on data analysis according to the present invention;

[0069] Figure 4 This is a schematic diagram of intelligent semantic annotation of a low-altitude airspace management method based on data analysis in the present invention. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0071] Example: Figure 1-Figure 4 As shown, the present invention provides a technical solution.

[0072] A low-altitude airspace management method based on data analysis, including the following contents:

[0073] The present invention uses dynamic grid division technology to adjust the grid granularity in real time according to the density of aircraft and environmental complexity to generate a multi-scale grid model; combined with intelligent airspace stratification technology, the thickness and distribution of airspace layers are dynamically adjusted through risk scoring and buffer layer optimization; in addition, intelligent semantic annotation technology is used to assign semantic information such as risk level and traffic priority to grid units. The present invention realizes the refined and intelligent management of low-altitude airspace and provides accurate flight planning support for aircraft in complex airspace environments; the above-mentioned content includes the following steps:

[0074] Step S1: Through the dynamic grid division method, combined with the aircraft density and environmental complexity data, the grid granularity is adjusted in real time to generate a multi-scale grid model. In order to improve the adaptability and efficiency of grid division, the optimized quadtree algorithm is used to generate the grid, and the grid boundary is smoothed by the Bezier curve interpolation method to ensure the continuity and consistency of the grid division. In areas with dense aircraft, the grid granularity is dynamically refined to improve navigation accuracy; in areas with simpler environments, the grid granularity is enlarged to optimize the computational burden. The grid division process can effectively reduce the risk of conflict and improve the efficiency of flight mission planning. In the end, the generated multi-scale grid can not only adapt to the complex airspace environment, but also lay an efficient data foundation for subsequent hierarchical airspace management and intelligent semantic labeling.

[0075] Step S2: Through intelligent airspace stratification technology, according to the aircraft density score, environmental complexity score and task priority score, combined with risk analysis methods, the thickness and distribution of the airspace layer are dynamically adjusted. Based on the risk score, the system can adaptively adjust the inter-layer structure, increase the number of layers and reduce the layer thickness in high-risk areas to provide more refined stratification support for aircraft; in low-risk areas, the number of layers is appropriately reduced and the inter-layer spacing is expanded, thereby optimizing the allocation efficiency of airspace resources. The buffer layer design further improves safety by dynamically calculating the buffer thickness to ensure that the aircraft transitions safely and reliably between adjacent layers. The overall stratification technology significantly improves resource utilization and flight safety, and is particularly suitable for the needs of airspace allocation flexibility and reliability in the multi-task execution environment of drones;

[0076] Step S3: Through intelligent semantic annotation technology, grid units are dynamically assigned risk levels, traffic priorities, and semantic information of regulations and rules to build an intelligent semantic grid. By parsing airspace control regulations, information on no-fly zones, restricted-fly zones, and recommended routes is embedded in the grid semantics, providing strong support for airspace management. In addition, through knowledge graphs and dynamic rule generation technology, semantic annotations can be updated according to real-time airspace environment changes to ensure the accuracy and timeliness of the annotation content. This intelligent semantic annotation technology not only improves the level of refinement of airspace management, but also significantly improves the execution efficiency of complex flight missions.

[0077] According to the above steps, the present invention further refines the multi-scale dynamic grid division operation and divides it into three specific steps: data collection and processing, dynamic grid size calculation and multi-scale grid generation.

[0078] The present invention first collects aircraft density and environmental complexity data in real time through multi-source sensor equipment, and uses noise filtering and standardization processing technology to generate high-quality dynamic scoring data. This operation ensures that the input data for dynamic mesh division is accurate and reliable. Secondly, through dynamic mesh size calculation, the present invention dynamically adjusts the mesh granularity according to the aircraft density and environmental complexity data. By using optimized formula calculation and adjustment mechanism, the adaptability and flexibility of the grid granularity are significantly improved, which is particularly suitable for dynamic airspace division in densely active areas and low-complexity scenes. Finally, the present invention uses an optimized quadtree algorithm to generate a multi-scale grid model, and further improves the continuity and inter-layer coordination of the grid through cross-layer topology construction and boundary smoothing interpolation. Through these refinement operations, the present invention can significantly improve the flexibility and accuracy of low-altitude airspace division, and provide more accurate navigation and flight planning support for aircraft.

[0079] According to the above content, step S1 includes the following sub-steps:

[0080] Step S11: Data collection and processing: collect aircraft density and environmental complexity data to generate dynamic scores;

[0081] Step S12: Dynamic grid size calculation: dynamically adjust the grid size according to the scoring data and generate a grid distribution map; the grid size calculation formula is as follows:

[0082]

[0083] Among them, S 基础 is the basic grid size, α and β are adjustment coefficients;

[0084] Step S13: multi-scale grid generation: generating a multi-scale grid model and optimizing its boundary continuity;

[0085] Step S14: Smoothing the boundaries of the multi-scale grids by using the Bezier curve interpolation method; ensuring that the boundaries at the junctions of different grid scales are continuous and consistent; the optimized grid boundaries are smoother, thereby improving the quality and adaptability of the overall grid model.

[0086] The present invention further refines data collection and processing and divides it into two specific steps: multi-source data collection and data preprocessing. The present invention first collects aircraft activity frequency, obstacle distribution and terrain complexity data in real time through multiple data sources (such as radar, optical sensors and drones). Combined with multi-source fusion technology, the present invention significantly improves the coverage and accuracy of data collection, and is particularly suitable for complex airspace scenarios. Subsequently, by performing noise filtering, outlier removal and feature extraction operations on the collected data, the present invention generates high-quality density scoring and complexity scoring data. In order to improve the adaptability of dynamic grid division, the present invention also introduces a dynamic normalization method, which can adjust the scoring range in real time according to regional characteristics. These refinement operations ensure the efficiency and reliability of dynamic grid division data input, and provide a solid data foundation for subsequent steps.

[0087] According to the content, step S11 has the following sub-steps:

[0088] Step S111: Multi-source data collection:

[0089] Step S111.1 Aircraft activity frequency data collection:

[0090] Data sources: Real-time collection of information such as the position, speed, heading, and activity frequency of aircraft through multiple data sources such as radar, optical sensors, and drones;

[0091] Data integration: Fusion of aircraft activity data from different data sources to ensure the comprehensiveness and accuracy of the data.

[0092] Step S111.2: Environmental complexity data collection:

[0093] The environmental complexity data mainly includes the following contents:

[0094] Topographic information:

[0095] Data sources: high-resolution topographic maps, geographic information system (GIS) data;

[0096] Content description: Includes the distribution and height information of terrain features such as mountains, rivers, urban buildings, bridges, etc.

[0097] Obstacle distribution data:

[0098] Data sources: LiDAR, drone aerial images, building sensor data;

[0099] Content description: Includes the number, location and dynamic changes of fixed obstacles (such as high-rise buildings and tower cranes) and dynamic obstacles (such as other aircraft and temporary structures).

[0100] Weather condition data:

[0101] Data sources: weather stations, satellite weather data, and meteorological sensors carried by drones;

[0102] Content description: Including meteorological parameters such as wind speed, wind direction, precipitation, visibility, temperature, etc. These parameters directly affect the flight safety and navigation accuracy of the aircraft.

[0103] Electromagnetic environment data:

[0104] Data sources: electromagnetic spectrum monitoring equipment, radio interference detection system;

[0105] Content description: Includes the distribution of electromagnetic interference sources in the area, electromagnetic wave intensity and frequency information to ensure the normal operation of aircraft communication and navigation systems.

[0106] Airspace usage data:

[0107] Data sources: airspace management system, flight plan database;

[0108] Content description: Includes information such as the frequency of airspace use in different time periods, the distribution of aircraft types and their flight altitudes, reflecting the utilization rate and potential complexity of airspace.

[0109] Step S112: Data preprocessing: noise filtering, outlier removal and standardization are performed on the collected data to generate density scores and complexity scores.

[0110] Step S113: Feature extraction: extract key feature indicators from the preprocessed data, including the aircraft movement speed and the environment change rate, to obtain the aircraft density score and the environment complexity score; wherein, the aircraft density score is calculated by a statistical method based on area division, the airspace is divided into several sub-areas, the number of aircraft appearances per unit time in each sub-area is counted, and then the density value is mapped to a score according to a preset scoring rule; the environment complexity score takes into account the terrain complexity, the number of obstacles and the dynamic obstacle factors, and uses a method based on multi-factor weighted average to calculate the environment change rate, and then according to the preset scoring rule, the environment change rate is mapped to a score.

[0111] The present invention further refines the operation of intelligent airspace stratification and divides it into three core steps: risk score calculation, dynamic stratification adjustment and buffer layer design. The present invention generates an airspace risk distribution map in real time according to the aircraft density, environmental complexity and task priority score through the risk score calculation method. This operation can accurately reflect the dynamic risk levels of different airspace areas and provide reliable data support for stratification adjustment. Dynamic stratification adjustment is combined with risk distribution maps to refine stratification for high-risk areas to improve the flexibility and safety of inter-layer distribution; stratification is simplified in low-risk areas to optimize resource allocation. The buffer layer design ensures the safe transition of aircraft between different levels by dynamically adjusting the buffer thickness, and is particularly suitable for conflict avoidance scenarios in complex flight environments. These refinement steps significantly improve the dynamic adaptability and refined management level of airspace stratification.

[0112] According to the content, step S2 includes the following sub-steps:

[0113] Step S21: Risk score calculation Calculate the comprehensive risk score and generate a risk distribution map; the risk score calculation formula is as follows: The formula is as follows:

[0114] Risk score = (w1×aircraft density score) + (w2×environmental complexity score) + (w3×mission priority score)

[0115] Among them, w1+w2+w3=1, and after calculation through the formula, the risk score data is obtained;

[0116] Step S22: Dynamic layer adjustment Adjust the airspace layer thickness according to the risk distribution; Dynamic adjustment of altitude layer thickness: Calculate the altitude layer thickness H according to the following formula 层厚 :

[0117] H 层厚 =H 基础 ×(1-(γ×risk score))

[0118] Among them, H 基础 is the thickness of the base layer, γ is the risk adjustment coefficient, and the dynamically adjusted height layer thickness data is obtained through calculation;

[0119] Step S23: Buffer layer design dynamically calculates the buffer layer thickness and optimizes the spatial inter-layer distribution; the buffer layer is set as follows: according to the aircraft speed V 飞行器 and safety time constant T 安全 To calculate the buffer layer thickness H 缓冲 , the formula is as follows:

[0120] H 缓冲 =V 飞行器 ×T 安全 .

[0121] The present invention further refines the risk score calculation operation and divides it into three core steps: weight allocation, score calculation and dynamic adjustment. The present invention first reasonably allocates the weights of density score, complexity score and priority score according to the mission requirements and airspace management needs. In the score calculation stage, a comprehensive risk score is generated by the weighted average method, and a risk distribution map is constructed to reflect the dynamic risk situation of the current airspace. Dynamic adjustment combines real-time data changes to optimize the score weights to improve the flexibility and adaptability of risk calculation. The entire operation ensures the accuracy and timeliness of the risk score, and provides a reliable basis for airspace stratification adjustment.

[0122] According to the above, step S21 further includes the following sub-steps:

[0123] Step S211: Weight allocation: reasonably allocate w1, w2 and w3 according to airspace management requirements and safety standards to obtain weight allocation data;

[0124] Step S212: Score calculation: use the weighted average method to calculate the comprehensive risk score and generate a risk distribution map;

[0125] Step S213: Dynamic adjustment: Optimize the scoring weights according to real-time data changes and dynamically update the risk score.

[0126] The present invention further refines the operation of dynamic stratification adjustment and divides it into three specific steps: layer thickness calculation, stratification optimization, and distribution verification. In the layer thickness calculation stage, the present invention dynamically generates thickness data of each layer according to the risk distribution map, and uses an optimization algorithm to adjust the inter-layer distribution to ensure that high-risk areas have more refined layer divisions. In the stratification optimization process, the stratification scheme is dynamically adjusted to adapt to different task requirements in combination with airspace utilization and safety requirements. Finally, the rationality and stability of the optimized stratification scheme are evaluated through the distribution verification method to ensure the scientificity and reliability of the stratification adjustment.

[0127] According to the above, step S22 includes the following sub-steps:

[0128] Step S221: Layer thickness calculation:

[0129] Step S221.1: Input data preparation:

[0130] Obtain risk distribution map data, including the risk score of each grid unit; obtain the thickness of the base layer H 基础 and the risk-adjusted coefficient γ.

[0131] Step S221.2: Layer thickness calculation formula application:

[0132] For each grid cell, according to its risk score R, the dynamically adjusted layer thickness H is calculated 层厚

[0133] H 层厚 =H 基础 ×(1-(γ×R))

[0134] Make sure that the calculated results comply with the preset minimum and maximum layer thickness limits to avoid over-adjustment.

[0135] Step S221.3: Data storage and update:

[0136] The calculated layer thickness data is stored in the airspace management database for subsequent stratification optimization and distribution verification; the layer thickness data is updated in real time to ensure the real-time and accuracy of dynamic adjustment.

[0137] Step S222: Layered optimization:

[0138] Step S222.1: Airspace demand analysis:

[0139] Analyze the mission requirements of the current airspace, including aircraft type, mission priority, and flight path planning; determine the functional division of airspace layers (such as priority passage layer, ordinary passage layer, etc.) based on the requirements of different tasks.

[0140] Step S222.2: Optimization algorithm application:

[0141] Multi-objective optimization algorithms (such as genetic algorithms and particle swarm optimization algorithms) are used to optimize the stratification scheme, with the goals of maximizing airspace utilization, minimizing the probability of aircraft conflicts and ensuring flight safety.

[0142] Set optimization constraints, such as the minimum safe distance between layers, upper and lower limits of layer thickness, etc.

[0143] Step S222.3: Layering scheme adjustment:

[0144] According to the optimization results, the distribution and thickness of each layer are dynamically adjusted to ensure that high-risk areas have more and thinner layers to improve stratification accuracy; for low-risk areas, the number of layers is appropriately reduced and the layer thickness is increased to improve computing efficiency and resource utilization.

[0145] Step S222.4: Feedback mechanism:

[0146] Establish a feedback mechanism to continuously optimize the stratification scheme based on actual flight data and stratification adjustment effects.

[0147] Step S223: Distribution verification:

[0148] Step S223.1: Verification index determination:

[0149] Determine the verification indicators of the stratification scheme, including stratification rationality (such as whether the layer thickness conforms to the risk distribution), stratification stability (such as the adaptability of the stratification scheme in a dynamic environment) and flight safety (such as the probability of aircraft collision).

[0150] Step S223.2: Simulation test:

[0151] Using flight simulation software, the optimized layering scheme was subjected to multi-scenario simulation testing to evaluate its performance under different aircraft densities and mission priorities.

[0152] Record the aircraft's flight path, conflict events, and inter-layer transitions, and analyze the rationality and stability of the layering scheme.

[0153] Step S223.3: Statistical analysis:

[0154] Conduct statistical analysis on the simulation test results, calculate key indicators such as aircraft conflict rate, flight path deviation rate, and inter-layer transition failure rate; compare the layering schemes before and after optimization to verify the improvement of the optimization effect.

[0155] The present invention further refines the buffer layer design and divides it into three sub-steps: buffer thickness calculation, conflict avoidance analysis, and inter-layer coordination optimization. In the buffer thickness calculation stage, the present invention dynamically generates buffer layer thickness data based on the aircraft speed and safety time constant to ensure safe transition between layers. Conflict avoidance analysis is combined with risk scoring to identify potential conflict points and optimize the buffer layer design. Inter-layer coordination optimization improves the safety and efficiency of coordinated operation of aircraft in high-density airspace environments by dynamically adjusting the buffer area.

[0156] According to the above, step S23 includes the following sub-steps:

[0157] Step S231: Calculation of buffer thickness:

[0158] Step S231.1: Input parameter acquisition:

[0159] Get the aircraft speed V 飞行器 and safety time constant T 安全 .

[0160] Step S231.2: Application of the calculation formula for buffer layer thickness:

[0161] Calculate the buffer layer thickness H 缓冲 :H 缓冲 =V 飞行器 ×T 安全 Ensure that the buffer layer thickness meets the minimum safety standards and adjust the calculation parameters according to different aircraft types.

[0162] Step S231.3: Data storage and update:

[0163] The calculated buffer layer thickness data is stored in the airspace management system for subsequent conflict avoidance analysis and inter-layer coordination optimization; and the buffer layer thickness data is updated in real time to adapt to changes in aircraft speed and safety time constant.

[0164] Step S232: conflict avoidance analysis, identifying potential conflicts and optimizing buffer layer design solutions;

[0165] Step S232.1: Identify potential conflict points

[0166] Data collection: Get the current flight path, speed and altitude information of the aircraft.

[0167] Conflict Detection: Using simple spatial and temporal analysis, identify areas where aircraft paths may cross or approach. For example, if two aircraft are likely to meet at the same altitude within the next 10 minutes, they are marked as potential conflict points.

[0168] Step S232.2: Evaluate conflict density

[0169] Conflict density calculation: Count the number of potential conflict points in each area. For example, one area has 3 potential conflict points and another area has 1.

[0170] Classification areas: The airspace is divided into high-risk areas and low-risk areas based on conflict density.

[0171] Step S232.3: Optimize buffer layer design

[0172] Adjust the buffer thickness:

[0173] High-risk areas: Increase the thickness of the buffer layer to provide a larger safety margin. For example, increase the buffer layer from 100 meters to 150 meters.

[0174] Low-risk areas: Reduce the thickness of the buffer layer to save airspace resources. For example, reduce the buffer layer from 100 meters to 80 meters.

[0175] Redistribute the buffer: Ensure that high-risk areas have adequate buffer coverage, while low-risk areas have a moderately reduced buffer.

[0176] Step S232.4: Implement the optimization plan

[0177] Update system configuration: Adjust the buffer layer parameters in the airspace management system based on the optimized buffer layer design.

[0178] Real-time monitoring and adjustment: Continuously monitor aircraft dynamics and dynamically adjust the thickness of the buffer layer when necessary to ensure response to new potential conflicts.

[0179] Step S233: Inter-layer coordination optimization:

[0180] Step S233.1: Dynamic adjustment of buffer area:

[0181] The distribution and thickness of the buffer zone are dynamically adjusted based on real-time flight data and conflict avoidance analysis results. In high-density airspace areas, the thickness of the buffer layer is appropriately increased to improve the safety of the transition between layers.

[0182] Step S233.2: Establishment of aircraft coordination operation mechanism:

[0183] Establish a communication coordination mechanism between aircraft to ensure that aircraft can obtain buffer layer adjustment information in a timely manner. Use a distributed coordination algorithm to optimize aircraft inter-layer switching and path planning to avoid inter-layer conflicts.

[0184] Step S233.3: Resource allocation optimization:

[0185] According to the buffer layer adjustment results, optimize the allocation of airspace resources, such as allocating more buffer layer resources to high-risk areas; adopt a dynamic resource scheduling algorithm to ensure efficient use and flexible allocation of buffer layer resources.

[0186] Step S233.4: Feedback and continuous optimization:

[0187] Collect data on inter-layer coordination operations, analyze the effects of buffer layer adjustments and aircraft coordination; based on the analysis results, continuously optimize the buffer layer design and inter-layer coordination mechanism to improve the efficiency and safety of overall airspace management.

[0188] The present invention further refines the operation of intelligent semantic annotation and divides it into three core steps: basic semantic assignment, regulatory information embedding and dynamic updating. The present invention first assigns basic semantic information such as risk level and traffic priority to grid units through dynamic annotation rules, and constructs preliminary grid semantic data in combination with airspace control requirements. Subsequently, by parsing airspace control regulations, information on no-fly zones, restricted-fly zones and recommended traffic paths is embedded into the grid model to provide comprehensive support for airspace management. Finally, using knowledge graph and rule engine technology, the system dynamically updates semantic information based on real-time airspace environment data to ensure the timeliness and accuracy of semantic annotation. These refinement steps have greatly improved the intelligence level of airspace management and the adaptability to complex tasks.

[0189] According to the above, step S3 includes the following sub-steps:

[0190] Step S31: basic semantic assignment: dynamically assigning risk level and traffic priority to grid cells;

[0191] Step S32: embedding regulatory information: embedding airspace control regulations into grid semantics;

[0192] Step S33: Dynamic update: Dynamically update semantic annotation information using knowledge graph and rule engine.

[0193] The present invention further refines the basic semantic assignment operation and divides it into three specific steps: dynamic rule generation, risk level labeling, and priority assignment. In the dynamic rule generation stage, the present invention combines real-time data and airspace mission requirements to generate semantic labeling rules adapted to the scenario. Risk level labeling combines aircraft density and environmental complexity scores to assign dynamic risk levels to grid cells to ensure the accuracy of risk distribution information. Priority assignment sets traffic priorities for grid cells in combination with mission requirements and control specifications, thereby optimizing the traffic efficiency and mission adaptability of the grid.

[0194] According to the above content, step S31 includes the following sub-steps:

[0195] Step S311: Dynamic rule generation, generating dynamic annotation rules according to real-time data and task requirements;

[0196] Step S311.1: Real-time data collection:

[0197] Data source: Obtain real-time aircraft position, speed, heading, mission type and other data, as well as environmental data such as weather conditions, terrain information, etc.

[0198] Data integration: Integrate multi-source data into a unified data platform to ensure the real-time and accuracy of the data.

[0199] Step S311.2: Task requirements analysis:

[0200] Mission classification: Classify according to the aircraft's mission type (such as patrol, transportation, monitoring, etc.).

[0201] Demand extraction: Extract the specific airspace requirements for different missions, such as right of way, specific flight altitude range, etc.

[0202] Step S311.3: Rule generation mechanism:

[0203] Rule template: Establish basic semantic annotation rule templates, including risk level classification standards and access priority setting rules.

[0204] Dynamic adjustment: Dynamically adjust rule parameters based on real-time data and mission requirements. For example, during periods of high aircraft density, increase the sensitivity of risk classification.

[0205] Step S312: risk level labeling, assigning risk levels to grid units in combination with scoring data;

[0206] S312.1: Scoring data acquisition:

[0207] Aircraft density score: Based on the aircraft density score data in step S1, it reflects the density of aircraft in the grid unit.

[0208] Environmental complexity score: Based on the environmental complexity score data, it reflects the complexity of the environment within the grid unit, such as the number of obstacles, terrain complexity, etc.

[0209] Step S312.2: Risk level classification:

[0210] Application of scoring weights: Based on the weight allocation (w1, w2, w3) in step S211, the aircraft density score and the environmental complexity score are weighted to obtain a comprehensive risk score.

[0211] Grading standards: Set risk score thresholds and divide the comprehensive risk score into different risk levels (such as low, medium, and high).

[0212] Step S312.4: Assigning semantic information:

[0213] Labeling operation: Assign a corresponding risk level mark to each grid unit according to the risk level.

[0214] Data storage: The annotated risk level information is stored in the semantic data of the grid model for use in subsequent steps.

[0215] Step S313: Assign priority values ​​and set the traffic priority according to task requirements and control specifications.

[0216] Step S313.1: Identification of task requirements:

[0217] Mission type classification: Identify the mission type and priority of the aircraft within the grid cell, such as emergency mission, high priority mission, etc.

[0218] Control specification review: Determine the traffic priority rules for different mission types based on airspace control regulations.

[0219] Step S313.2: Priority setting:

[0220] Priority rule application: According to the dynamically generated rules (step S311), the aircraft in the grid cell are assigned a priority. For example, a rescue mission has a higher priority than a normal transport mission.

[0221] Priority level classification: Set the level of access priority (such as high, medium, and low) and divide it according to mission requirements and control regulations.

[0222] Step S313.3: semantic information integration:

[0223] Labeling operation: Assign a corresponding traffic priority mark to each grid unit.

[0224] Data update: Integrate priority information with risk level information to form complete grid semantic data.

[0225] The present invention further refines the operation of embedding regulatory information and divides it into three specific steps: regulatory parsing, semantic model generation, and rule mapping. The present invention first converts airspace control regulations into structured semantic rule data through a regulatory parsing module. Semantic model generation combines regulatory data and grid characteristics to build a multi-level semantic model to provide semantic support for complex airspace management. Rule mapping dynamically matches the parsed regulations with grid units to achieve a deep fusion of regulatory information and airspace semantics, thereby improving the accuracy and compliance of semantic annotation.

[0226] According to the above content, the S32 includes the following sub-steps:

[0227] Step S321: Regulation parsing, parsing airspace control regulations and generating structured semantic rule data;

[0228] Step S322: semantic model generation, combining regulatory data and grid characteristics to generate a multi-level semantic model;

[0229] Step S323: Rule mapping, dynamically mapping regulatory data to grid units to achieve semantic fusion.

[0230] The present invention further refines the dynamic update operation and divides it into three specific steps: real-time data collection, knowledge graph construction, and rule engine optimization. In the real-time data collection stage, the present invention combines multi-source sensors to collect information such as aircraft activity frequency and airspace environment data to provide real-time input for dynamic updates. Knowledge graph construction constructs a dynamic knowledge graph by associating semantic models and real-time data for contextual reasoning and supplementation of semantic information. Rule engine optimization combines real-time updated data and rule sets to dynamically generate or adjust semantic rules to ensure the timeliness and consistency of airspace semantic annotation.

[0231] According to the above content, step S33 includes the following sub-steps:

[0232] Step S331: real-time data collection, collecting information such as aircraft activity frequency, airspace environment data, etc.;

[0233] Step S332: knowledge graph construction, associating the semantic model with real-time data to construct a dynamic knowledge graph;

[0234] Step S333: Optimize the rule engine, dynamically generate or adjust semantic rules, and ensure the timeliness of semantic annotation.

[0235] To achieve the above object, the present invention provides the following technical solutions:

[0236] A low-altitude airspace management system based on data analysis, including the following contents:

[0237] The system includes a data processing and grid generation module, an intelligent airspace stratification module, a semantics assignment and update module, and a control and decision module;

[0238] The data processing and grid generation module is used to execute the multi-scale dynamic grid division step, responsible for the collection and processing of real-time data and the generation and optimization of multi-scale grids, and obtains multi-scale grid data and grid boundary data after boundary smoothing;

[0239] The intelligent airspace stratification module is used to execute the intelligent airspace stratification steps, is responsible for risk score calculation, altitude layer thickness adjustment and buffer layer setting, realizes dynamic stratification management of airspace, and obtains an intelligent stratified airspace model;

[0240] The semantic assignment and update module is used to execute the intelligent semantic annotation step, responsible for semantic information annotation, rule embedding and real-time update of semantic information of grid units, and obtains real-time updated semantic annotation data;

[0241] The control and decision module is used to make airspace management decisions and controls based on multi-scale grid data, intelligently layered airspace models and real-time updated semantic annotation data to ensure the normal operation of aircraft in low-altitude airspace.

[0242] Assume that there is a busy low-altitude airspace, which covers part of the suburban area (with undulating terrain, buildings, high-voltage towers and other obstacles) and an adjacent open plain for drone logistics distribution and general aviation flight training. The airspace is frequently used for daily aircraft activities, including drones of different types and mission priorities, as well as small private aircraft.

[0243] 1. Multi-source data collection:

[0244] Radar equipment is rationally arranged around this low-altitude airspace to monitor the position, speed and other information of aircraft in a larger range, so as to count the frequency of aircraft activities; optical sensors are installed at key high points (such as mountaintop observation stations and the tops of tall buildings) to perform high-precision imaging of the terrain and obstacles below the airspace and accurately identify the distribution of obstacles; at the same time, multiple drones with data collection functions are dispatched to shuttle between different altitude layers in the airspace according to preset routes to assist in collecting more detailed local environmental data;

[0245] Through data collection over a period of time (assuming 1 hour), the radar records that the number of aircraft passing through the airspace at different times fluctuates greatly, with an average of 5-10 aircraft passing through the specific monitoring area every 10 minutes, reflecting a high frequency of activity; optical sensor imaging clearly depicts the exact location and outline of more than 50 obstacles; drone data collection supplements the complex terrain details of some hidden areas such as valleys and woods.

[0246] 2. Data preprocessing:

[0247] The collected data is filtered for noise, and abnormal data such as false echo signals generated by electromagnetic interference of radar and imaging noise of optical sensors are removed; outlier data points due to positioning deviation during drone collection are eliminated. Feature extraction is then performed, and the peak value, mean value and other features of the aircraft activity frequency are statistically converted into aircraft density scores by time period (every 10 minutes). The value range is set to 0-10 (0 represents extremely low density, 10 represents extremely high density), and the average score of the current area is calculated to be 6. A comprehensive assessment of the terrain complexity and obstacle distribution is performed, and the environmental complexity score is converted according to the complexity level. The value is also 0-10. Here, due to the undulating terrain and more obstacles, the score is 7. A dynamic normalization method is introduced. According to the statistical characteristics of the past data in the area, the score range is appropriately scaled to make it more in line with the local actual situation, and finally high-quality density score and complexity score data are obtained for subsequent grid division.

[0248] 3. Dynamic grid size calculation:

[0249] Set the base grid size S 基础 For 1000 meters, the adjustment coefficient α = 0.02, β = 0.1, using the formula

[0250]

[0251] To calculate the grid granularity, substitute the previously obtained aircraft density score 6 and environment complexity score 7, and we get:

[0252] S=1000×(1-(0.02×6+0.1×7))

[0253] S=180

[0254] According to the calculation results, it is shown that the grid size planning is reasonable. Then, the optimized quadtree algorithm is used to divide the airspace layer by layer according to the grid granularity calculated above to construct a multi-scale grid model. In the division process, attention is paid to cross-layer topology construction to ensure that the grids at different levels are connected reasonably and the topological relationship is clear, which is convenient for data association and management. After that, the Bezier curve interpolation method is used to smooth the grid boundaries, especially in the junction area of ​​grids of different scales. After smoothing, the originally jagged and rigidly spliced ​​boundaries become continuous and smooth, which improves the overall grid model quality and makes it more in line with the actual geographical and flight environment characteristics of the airspace, laying a solid foundation for subsequent airspace stratification and semantic labeling.

[0255] 4. Risk score calculation:

[0256] Weight allocation: According to the characteristics of the activities such as logistics distribution (higher mission priority) and training flight (medium mission priority) in this airspace and the safety management needs, the weights w1 (focusing on the impact of aircraft density on risk, aircraft are densely populated in logistics distribution concentrated areas), w2 (environmental complexity is related to flight safety, and areas with obstacles and undulating terrain have high risks), and w3 (mission priorities distinguish the urgency and importance of different flight missions);

[0257] Rating calculation: Given the aircraft density score of 6, the environmental complexity score of 7, and the mission priority score (logistics mission is 8, training mission is 6, and the comprehensive average is 7), the risk score calculation formula is used.

[0258] Risk score = (w1×aircraft density score) + (w2×environmental complexity score) + (w3×task priority score); Substituting the data into the risk score, the risk score is 6.6; then the airspace risk distribution map is drawn based on this, and the high-risk areas are concentrated above the logistics distribution center where aircraft frequently take off and land, and in valleys with complex terrain and many obstacles.

[0259] Dynamic adjustment: As the peak period of logistics and distribution in the airspace ends, the density of aircraft decreases (the new score becomes 4), the environmental complexity remains unchanged, and the overall mission priority decreases slightly (the new score is 6). The risk score is recalculated: Substituting the data, the risk score is 5.5, and then the risk score is updated in a timely manner to dynamically reflect changes in airspace risks.

[0260] 5. Dynamic layering adjustment:

[0261] Layer thickness calculation: Assume that the base layer thickness H 基础 = 100 meters, risk adjustment coefficient γ = 0.2. The risk score of a grid unit R = 0.75, then the layer thickness after dynamic adjustment is:

[0262] H layer thickness = 100 × (1-(0.2 × 0.75)) = 100 × (1-0.15) = 85 meters

[0263] If R = 0.4, then:

[0264] H layer thickness = 100×(1-(0.2×0.4)) = 100×0.92 = 92 meters.

[0265] Layered optimization: In a complex airspace area, the optimization algorithm determines that the area needs to be divided into three layers: First, the traffic layer: 80 meters thick, mainly serving high-priority mission aircraft. Ordinary traffic layer: 100 meters thick, serving ordinary mission aircraft. Buffer layer: 50 meters thick, to ensure the safety of the transition between layers.

[0266] 6. Buffer layer design:

[0267] Buffer thickness calculation: In the simulation test, the optimized layering scheme was applied to simulate 1,000 flight missions, and the results showed that the aircraft conflict rate was reduced by 30%, the inter-layer transition failure rate was reduced to less than 1%, and the flight path deviation rate was reduced by 15%.

[0268] Expert review feedback indicated that the stratified scheme has good adaptability and efficiency in complex airspace environments, and it is recommended to further optimize the buffer layer design to improve transition safety. Based on the feedback, the calculation formula for the buffer layer thickness was adjusted to further improve the stability of the stratified scheme.

[0269] Assume the aircraft speed V 飞行器 =200 m / s, safety time constant T 安全 = 0.5 seconds, then the thickness of the buffer layer is: H 缓冲 =200×0.5=100 meters

[0270] If the aircraft speed increases to V 飞行器 =250 m / s, the thickness of the buffer layer is:

[0271] H 缓冲 =250×0.5=125 meters.

[0272] Conflict avoidance analysis: Combined with risk scoring, potential conflict points are identified in areas with higher risks and a high possibility of aircraft trajectory intersection (such as the intersection of take-off and landing routes in logistics centers), the buffer layer is appropriately thickened to 300 meters and the design plan is optimized, and the shape of the buffer layer and guidance rules are adjusted.

[0273] Inter-layer coordination optimization: Assume that two high-risk areas and one low-risk area are identified in the current airspace: High-risk area A: There are 2 potential conflict points; optimization measures: Increase the thickness of the buffer layer from 100 meters to 150 meters. High-risk area B: There is 1 potential conflict point; optimization measures: Increase the thickness of the buffer layer from 100 meters to 130 meters. Low-risk area C: No potential conflict point; optimization measures: Reduce the thickness of the buffer layer from 100 meters to 80 meters.

[0274] 7. Basic semantic assignment:

[0275] Dynamic rule generation: Generate semantic annotation rules based on real-time data (current aircraft density, mission type ratio, etc.) and mission requirements (ensuring the timeliness of logistics distribution and the standardization of training flights). For example, high-pass priority rules are given to high-frequency path grids for logistics drones, and high-risk level rules are set for low-altitude grids in complex terrain.

[0276] Risk level annotation: The aircraft density score of a grid unit is 0.6, and the environmental complexity score is 0.4. The weight distribution is w1 = 0.5, w2 = 0.3, w3 = 0.2 (the task priority score is not considered for the time being). The comprehensive risk score is calculated as follows: Risk score = (0.5 × 0.6) + (0.3 × 0.4) = 0.3 + 0.12 = 0.42

[0277] According to the set threshold (such as 0-0.3 for low risk, 0.3-0.6 for medium risk, and 0.6-1 for high risk), the grid cell is assigned a "medium risk" level.

[0278] Priority assignment: Assuming that there are multiple drones performing different tasks, the system generates the following dynamic rules based on real-time data: High-density period: The risk level classification standard is raised, that is, in areas with dense aircraft, lower aircraft density will also be marked as high risk. Specific mission priority: Aircraft with transportation missions have higher priority than aircraft with monitoring missions.

[0279] 8. Regulatory information embedding:

[0280] Regulation analysis: The regulation analysis module interprets local low-altitude airspace control regulations and converts the no-fly zone range, restricted-fly zone altitude and time period restrictions into structured semantic rule data;

[0281] Semantic model generation: Combine regulatory data and grid characteristics (such as grid geographic location, size, and surrounding environment) to build a multi-level semantic model, assign absolute no-passage semantic labels to no-fly zone grids, and assign time-limited and highly restricted semantic descriptions to restricted flight zones according to restrictions, providing precise semantic support for complex airspace management;

[0282] Rule mapping: Dynamically match the parsed regulations with grid units, deeply integrate regulatory semantic information into the grid, ensure that no-fly and restricted-fly semantics are accurately associated with corresponding airspace grids, and ensure labeling compliance.

[0283] 9. Dynamic Update:

[0284] Real-time data collection: Continuously use multi-source sensors to collect information such as aircraft activity frequency (a sudden increase in aircraft near a newly opened temporary take-off and landing point) and airspace environment data (local terrain occlusion due to the construction of new high-rise buildings) to provide real-time input for semantic updates;

[0285] Knowledge graph construction: associate semantic models with real-time data to build a dynamic knowledge graph, such as associating the location of a newly built high-rise building with the surrounding grid risk level enhancement semantics, and associating busy take-off and landing points with high-speed traffic priority semantics, to supplement semantic reasoning and improve annotation logic;

[0286] Rule engine optimization: Combine new data and rule sets to dynamically adjust semantic rules, update risk annotations for grids around newly built high-rise buildings, and reset traffic priorities near take-off and landing points to ensure the timeliness and consistency of semantic annotations to keep up with real-time changes in airspace.

[0287] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A low-altitude airspace management method based on data analysis, characterized in that: The following steps are involved: Step S1: Multi-scale dynamic grid division: By acquiring real-time regional characteristic data, aircraft density data and environmental complexity data are obtained; aircraft density data and environmental complexity data are standardized to obtain aircraft density scores and environmental complexity scores; grid sizes are calculated based on aircraft density scores and environmental complexity scores to obtain dynamic grid sizes; multi-scale grids are generated using an optimized quadtree data structure to obtain multi-scale grid data; the boundaries of multi-scale grids are smoothed using the Bezier curve interpolation method to obtain grid boundary data with smooth transitions; Step S2: Intelligent airspace stratification: Calculate the risk score according to the aircraft density score, environmental complexity score and task priority score to obtain the risk score data; calculate the altitude layer thickness according to the risk score data to obtain the dynamically adjusted altitude layer thickness; calculate the buffer layer thickness according to the aircraft speed and the safety time constant to obtain the buffer layer thickness data; perform airspace stratification according to the risk score data and the buffer layer thickness data to obtain an intelligent stratified airspace model; Step S3: Intelligent semantic annotation: The grid units are labeled with risk levels and traffic priorities according to dynamic labeling rules to obtain preliminary semantic labeling data; grid semantics are embedded according to information on airspace control regulations, no-fly zones, restricted-fly zones and recommended flight channels to obtain semantic labeling data with embedded regulatory information; semantic information is updated in real time through knowledge graphs and rule engines to obtain real-time updated semantic labeling data.

2. A low-altitude airspace management method based on data analysis according to claim 1, characterized in that: The multi-scale dynamic grid division comprises the following steps: Step S11: Real-time regional characteristic evaluation: obtain aircraft density data and environmental complexity data through the actual regional characteristic data obtained; perform standardization processing on the aircraft density data and environmental complexity data to obtain an aircraft density score and an environmental complexity score; Step S12: Dynamic grid size calculation: According to the aircraft density score and the environmental complexity score, the grid size S is calculated using the following formula: 网格 : Among them, S 基础 is the basic grid size, α and β are adjustment coefficients; Step S13: generating a multi-scale grid: generating a multi-scale grid using an optimized quadtree data structure according to the dynamically adjusted grid size data, and obtaining multi-scale grid data; Step S14: Boundary smoothing: Based on the multi-scale grid data, a Bezier curve interpolation method is used to perform smooth transition processing on the boundary areas of different grid scales to obtain smooth grid boundary data with continuous and consistent boundaries.

3. A low-altitude airspace management method based on data analysis according to claim 2, characterized in that: The real-time regional characteristics assessment further includes the following: Step S111: Data collection: collecting aircraft density data and environmental complexity data in real time through multi-source sensors; Step S112: Data preprocessing: noise filtering, outlier processing and data cleaning are performed on the collected data to obtain high-quality preprocessed data; Step S113: Feature extraction: extract key feature indicators from the preprocessed data, including the aircraft movement speed and the environment change rate, to obtain the aircraft density score and the environment complexity score; wherein, the aircraft density score is calculated by a statistical method based on area division, the airspace is divided into several sub-areas, the number of aircraft appearances per unit time in each sub-area is counted, and then the density value is mapped to a score according to a preset scoring rule; the environment complexity score takes into account the terrain complexity, the number of obstacles and the dynamic obstacle factors, and uses a method based on multi-factor weighted average to calculate the environment change rate, and then according to the preset scoring rule, the environment change rate is mapped to a score.

4. A low-altitude airspace management method based on data analysis according to claim 2, characterized in that: The dynamic grid size calculation further includes the following: Step S121: Adaptive adjustment mechanism: dynamically adjust the adjustment coefficients α and β according to the real-time regional characteristic evaluation results and system load conditions to obtain optimized adjustment coefficients; Step S122: Error correction: performing error detection on the calculated grid size, and using an error correction algorithm to correct the grid size to obtain corrected grid size data; Step S123: Multi-level optimization algorithm: combining global optimization and local optimization strategies, a multi-level optimization algorithm is used to optimize the grid division to obtain optimized multi-scale grid data.

5. The low-altitude airspace management method based on data analysis according to claim 1 is characterized in that: The intelligent airspace layering further includes the following contents: Step S21: Risk score calculation: The risk score is calculated by integrating the aircraft density score, the environmental complexity score, and the task priority score according to the weights w1, w2, and w3. The formula is as follows: Risk score = (w1×aircraft density score) + (w2×environmental complexity score) + (w3×mission priority score) Among them, w1+w2+w3=1, and after calculation through the formula, the risk score data is obtained; Step S22: Dynamic adjustment of height layer thickness: Calculate the height layer thickness H according to the following formula: 层厚 : H 层厚 =H 基础 ×(1-(γ×risk score)) Among them, H 基础 is the thickness of the base layer, γ is the risk adjustment coefficient, and the dynamically adjusted height layer thickness data is obtained through calculation; Step S23: Buffer layer setting: according to the aircraft speed V 飞行器 and safety time constant T 安全 To calculate the buffer layer thickness H 缓冲 , the formula is as follows: H 缓冲 =V 飞行器 ×T 安全 。 6. A low-altitude airspace management method based on data analysis according to claim 5, characterized in that: The risk score calculation further includes the following contents: Step S211: Weight allocation: According to the management requirements and safety standards of different regions, w1, w2 and w3 are reasonably allocated to obtain weight allocation data; Step S212: weighted average method: the weighted average method is used to combine the aircraft density score, environmental complexity and task priority score to obtain a comprehensive risk score; Step S213: Dynamic adjustment: Dynamically adjust the weight distribution according to the real-time airspace environment and flight mission requirements, optimize the flexibility and adaptability of the risk score, and obtain the adjusted risk score data.

7. The low-altitude airspace management method based on data analysis according to claim 1 is characterized in that: The intelligent semantic annotation further includes the following contents: Step S31: Dynamic labeling rules: assign risk level and traffic priority information to grid cells, and adjust labeling rules based on real-time data; Step S32: semantic rule embedding: embed the information of airspace control regulations, no-fly zones, restricted-fly zones and recommended flight channels into the grid semantics, and update them in real time using the knowledge graph and rule engine.

8. A low-altitude airspace management method based on data analysis according to claim 7, characterized in that: The semantic rule embedding further includes the following contents: Step S321: Regulation parsing: parsing the latest airspace control regulations and converting them into semantic rules that can be processed by the system to obtain parsed regulation data; Step S322: knowledge graph construction: by constructing a knowledge graph based on the parsed regulations and real-time data, establishing the association between airspace areas and aircraft to obtain knowledge graph data; Step S323: Rule engine generation: Dynamically generate and update semantic rules based on the knowledge graph and real-time data to ensure the accuracy of semantic information and obtain real-time updated semantic rule data. Step S324: semantic verification: verify and correct the embedded semantic information, use a verification algorithm to ensure the correctness and consistency of the semantic annotation, and obtain verified semantic annotation data.

9. A low-altitude airspace management system based on data analysis, such as a low-altitude airspace management method based on data analysis as described in any one of claims 1 to 8, characterized in that: It includes data processing and grid generation module, intelligent airspace stratification module, semantic assignment and update module and control and decision module; The data processing and grid generation module is used to execute the multi-scale dynamic grid division step, responsible for the collection and processing of real-time data and the generation and optimization of multi-scale grids, and obtains multi-scale grid data and grid boundary data after boundary smoothing; The intelligent airspace stratification module is used to execute the intelligent airspace stratification steps, is responsible for risk score calculation, altitude layer thickness adjustment and buffer layer setting, realizes dynamic stratification management of airspace, and obtains an intelligent stratified airspace model; The semantic assignment and update module is used to execute the intelligent semantic annotation step, responsible for semantic information annotation, rule embedding and real-time update of semantic information of grid units, and obtains real-time updated semantic annotation data; The control and decision module is used to make airspace management decisions and controls based on multi-scale grid data, intelligently layered airspace models and real-time updated semantic annotation data to ensure the normal operation of aircraft in low-altitude airspace.

Images