Urban airspace evaluation method based on availability quantification and multi-risk evaluation

Abstract

The invention discloses an urban airspace evaluation method based on availability quantification and multi-risk evaluation, and relates to the technical field of urban airspace evaluation, and the method is characterized in that the method comprises the following steps: S1, airspace availability modeling: carrying out the three-dimensional discretization of an urban low-altitude region; s2, performing a static risk model, wherein a final static comprehensive risk value is obtained by performing scientific weight distribution on four types of risks according to an entropy weight method; s3, a dynamic risk model which can reflect personnel activity rules of the city in different time periods and realize time-varying risk assessment of the city low-altitude airspace; and S4, carrying out eight-quadrant analysis and improved Pareto sorting. The technical problem to be solved by the invention is to provide the urban airspace evaluation method based on availability quantification and multi-risk evaluation, the precision, credibility and application value of urban low-altitude airspace evaluation can be significantly improved, and a scientific basis and technical support are provided for safe operation of an unmanned aerial vehicle, airspace planning and air route design.

Description

Technical Field

[0001] This invention relates to the field of urban airspace assessment technology, and more specifically, to an urban airspace assessment method based on availability quantification and multi-risk assessment. Background Technology

[0002] Current research on urban low-altitude airspace mainly revolves around airspace assessment, modeling, and delineation. In airspace modeling, some studies utilize real urban building data to construct low-resolution 3D models and evaluate usability through indicators such as airspace accessibility and traffic flow. However, due to their coarse-grained representation, these models are insufficient in depicting changes in spatial connectivity caused by building obstruction, making it difficult to accurately reflect the real airspace feasibility risks under complex urban structures. Regarding risk modeling, existing research focuses more on third-party risks such as population mortality risk, property damage risk, and noise impact risk, while research on LOS signal transmission risk, building collision risk, and indirect mortality risk is insufficient. Furthermore, current research on noise impact risk largely remains theoretical, rarely or almost never linking the noise generated by drones to population distribution or high-precision urban GIS data. Although some works attempt to incorporate changes in population density or traffic volume to assess direct ground-based casualty risks, they mostly employ static or time-segmented statistical methods, lacking a dynamic spatiotemporal representation of population flow characteristics and failing to reflect the true characteristics of risk changes over time.

[0003] Furthermore, under multi-indicator conditions, most existing airspace assessment methods are single-indicator, with very few studies employing dual-indicator assessments. These methods primarily consider factors such as population risk and airspace availability, leading to significant randomness in airspace assessment. When conflicts arise between different indicators, the stability and applicability of such methods decrease significantly. Although some studies have introduced multi-objective optimization or improved Pareto ranking to enhance airspace classification quality, insufficient risk type coverage, limited input data accuracy, and constrained modeling capabilities result in airspace maps with low resolution and weak operability, making them difficult to directly apply to UAV operational mission planning.

[0004] Overall, existing research mostly focuses on single or dual indicators, lacking comprehensive modeling and systematic analysis of multiple risk factors, and the accuracy of airspace delineation is generally low. Most studies have not adequately addressed the multi-dimensional safety factors affecting UAV operation in complex urban environments, resulting in significant deficiencies in the applicability and practical value of their feasibility assessments.

[0005] Existing technologies for assessing and ensuring the safe operation of urban low-altitude airspace still have significant shortcomings in real-world urban scenarios. On the one hand, existing airspace modeling methods often rely on idealized or simplified environmental representations, such as two-dimensional ground projection models, low-precision grid models, or simplified building structure models. These methods struggle to accurately reflect changes in building density, weakened spatial connectivity, and morphological heterogeneity in complex urban environments, thus limiting the accuracy and reliability of airspace availability assessments. On the other hand, urban low-altitude operations involve multiple risk sources, including LOS signal transmission obstruction, obstacle collision risks, direct and indirect injury risks, property damage risks, and social noise impacts. Existing studies typically only consider mortality risks, neglecting other risks and lacking a unified multi-source risk modeling mechanism, thus failing to comprehensively reflect the true safety level of urban low-altitude airspace operations.

[0006] Furthermore, urban population and traffic exhibit significant temporal variations, but existing studies generally use static population density or fixed traffic conditions to represent risk, failing to reflect the differences in risk characteristics under time-varying environments. This makes airspace delineation unable to adapt to operational needs across different time periods. In terms of multi-indicator fusion and decision-making, traditional methods often employ fixed thresholds or linear weighting, making it difficult to achieve reasonable ranking under conflicting multi-indicator conditions, resulting in insufficient stability of evaluation results. Simultaneously, for large-scale, high-resolution 3D airspace modeling tasks, existing methods are insufficient in terms of computational feasibility and execution efficiency, making it difficult to support the operational management needs of unmanned aerial vehicles (UAVs). Summary of the Invention

[0007] The technical problem to be solved by this invention is to provide an urban airspace evaluation method based on availability quantification and multi-risk assessment, which can significantly improve the accuracy, reliability and application value of urban low-altitude airspace assessment, and provide scientific basis and technical support for the safe operation of UAVs, airspace planning and route design.

[0008] The present invention achieves its objective by employing the following technical solution: A method for evaluating urban airspace based on availability quantification and multi-risk assessment, characterized by the following steps: S1: Airspace availability modeling, which performs three-dimensional discretization of the urban low-altitude area, divides the study area into several voxel grids for fine characterization of the urban low-altitude spatial structure; based on actual geographic information system building data, the height and location of buildings are mapped to corresponding voxel units, and obstacles are marked on the voxels occupied by buildings. S2: Static risk model, used to assess risk factors in urban low-altitude airspace that do not change over time, mainly including line-of-sight signal transmission risk, obstacle collision risk, indirect injury risk and property damage risk. The final static comprehensive risk value is obtained by scientifically allocating weights to the four types of risks using the entropy weight method. S3: Dynamic risk model, which can reflect the patterns of human activity in the city at different times and realize time-varying risk assessment of urban low-altitude airspace; S4: Eight-quadrant analysis and improved Pareto ranking method. Based on airspace availability, static risk and dynamic risk, a four-quadrant classification system is constructed that combines three indicators. The three indicators are divided into high and low levels, forming a total of eight combinations. Based on quadrant classification, to further improve the accuracy and efficiency of spatial domain screening, an improved Pareto ranking method is introduced. By constructing a linear objective function to solve the three types of indicators in a weighted manner, a comprehensive ranking of voxel units is achieved.

[0009] As a further limitation of this technical solution, step S1 is as follows: S11: Construct a 5×5×5 (k=2) k-cuboid region centered on the evaluation voxel as the spatial analysis unit, and calculate the minimum hop count relationship between voxels; from voxels i Totozotoxin j The minimum number of hops is d unobs ( i , j ); S12: In a real airspace containing static obstacles such as buildings, the minimum number of hops is denoted as... d obs ( i , j Due to the bypass effect, and satisfying d obs ( i , j )≥ d unobs ( i , j ); S13: This invention constructs a spatial coverage index: (1); The overall voxel availability index is obtained by averaging all available voxels within the k-cuboid region. (2); in: Indicated by voxels The effective voxel set within the k-cuboid coverage area centered on the core. Effective voxels are voxel units not occupied by building obstacles. | represents the number of voxels in the set.

[0010] As a further limitation of this technical solution, S2 specifically includes: S21: Risks associated with line-of-sight signal transmission; Based on whether there is a Line-of-Sight (LOS) link between the antenna location and other grid cells, the communication risk is 0 when LOS exists; if the link is obstructed by a building, it is considered a communication interruption with a risk of 1. The mathematical model is as follows: (3); in; R 1 indicates a risk index for LOS signal transmission; Indicates "existence"; Indicates "does not exist"; LOS represents the unobstructed line-of-sight propagation path between the drone and the communication node; The communication risk is lowest when LOS exists and highest when LOS does not exist. S22: Risk of collision with obstacles; The closer the drone is to the obstacle, the higher the risk of collision; to avoid unlimited risk decay, a threshold is set. min If the distance between the drone and the target is greater than the threshold, it is considered to have no collision risk; if the distance is 0, the risk is recorded as 1. The obstacle collision risk is defined as follows: (4); in: R 2 indicates an obstacle collision risk index; dist_obs The minimum distance between the drone and city buildings; k GNSS For global satellite navigation system accuracy adjustment factors; S23: Risk of indirect injury or death; The risk model is as follows: (5); in: R 3 indicates an indirect injury risk indicator; P UCO This represents the probability of a building collision. N b_impact The number of buildings affected by the drone crash; N D P This is the average number of casualties coefficient; S24: Risk of property loss; Based on the type of ground object, three loss scenarios are defined: (6); Property damage risk is calculated using the loss coefficient, collision kinetic energy, and system failure probability. (7); in: R 4 indicates a risk indicator of property loss; P crash This represents the failure probability of the unmanned aerial vehicle (UAV) system. E impact This refers to the actual collision energy. E max The maximum possible collision energy; S25: Static Risk Comprehensive Model; The weights are automatically calculated using the entropy weight method. r i Its general formula is: (8); in: Indicates the first Weighting coefficients for static risk indicators; Indicates the first The information utility value of risk indicators; Indicates the first Information entropy of risk indicators; Indicates the first Individual unit in the first Normalized risk value under similar risk indicators; This represents the total number of voxel units involved in the risk calculation. It is a very small positive number, used to avoid numerical instability in logarithmic calculations; The final static risk model is: (9).

[0011] As a further limitation of this technical solution, the specific steps of S3 are as follows: S31: In order to obtain the input parameters of the dynamic risk model, the urban population density is extrapolated in a secondary distribution based on land use type, so as to construct the ground population density model for different time periods. S32: Construction of dynamic risk indicators; Dynamic risks vary depending on population distribution at different times, including the direct risk of injury or death to pedestrians or vehicles caused by drone crashes. R 5. Social impact risks of drone noise on the ground population RThe comprehensive assessment value of dynamic risk, consisting of 6 components, is defined as follows: (10); in: w 5. w 6 represents the weights of the two types of risks, satisfying... w 5+ w 6 = 1; The risk of direct death is due to the risk of injury or death to pedestrians. R fat_p Risk of vehicle injury or death R fat_v The weighted combination consists of: (11); pedestrian injury and death risk R fat_p Represented as: (12); in: P crash The probability of a drone crashing due to system malfunction; N hit p The number of pedestrians within the area affected by the fall; R f P The fatality rate of pedestrians after being hit; Vehicle injury risk R fat_v Represented as: (13); in: N hit v Vehicles within the impact area of ​​the fall; R f v The fatality rate of pedestrians after being hit; Social impact risk is used to measure the disturbance caused by drone noise in densely populated areas on the ground. Its calculation formula is as follows: (14); in: L h For reference noise intensity; ω This is the conversion factor from noise to impact level; H This refers to the flight altitude of the drone. d noiseThe horizontal distance between the drone and the center of the affected grid.

[0012] As a further limitation of this technical solution, in S4, the objective function of the improved Pareto sorting method is: (15); in: f 1、 f 2 represents the normalized value of static risk and dynamic risk; α This is an indicator of airspace availability. w 1, w 2, w 3 represents the weighting coefficient.

[0013] Compared with the prior art, the advantages and positive effects of the present invention are: 1. This invention, based on high-precision urban GIS data, introduces a quantitative index for airspace availability based on the shortest hop count and integrates a risk assessment framework combining static and dynamic risks. It achieves unified modeling of risks such as LOS signal transmission risk, building collision risk, direct and indirect fatality risk, population and traffic distribution, property damage risk, and social impact risk. Furthermore, this invention combines eight-quadrant analysis and an improved Pareto ranking strategy to construct a multi-indicator comprehensive evaluation mechanism, which can generate hierarchical, executable, and dynamically adjustable feasible airspace structures in large-scale urban airspace. This invention can significantly improve the accuracy, reliability, and application value of urban low-altitude airspace assessment, providing scientific basis and technical support for the safe operation of UAVs, airspace planning, and route design.

[0014] 2. This invention has significant advantages over existing technologies. First, existing methods often employ low-resolution modeling, which cannot accurately reflect building obstruction and spatial structure changes in urban environments. This invention, however, uses high-resolution urban GIS data modeling, ensuring high accuracy in airspace availability assessment. Second, existing technologies primarily focus on third-party risks such as population mortality risk, property damage risk, and noise impact risk, neglecting other risks. This invention, by integrating static and dynamic risk models, comprehensively assesses multiple risk factors such as LOS signal transmission, obstacle collisions, and social impact, improving the comprehensiveness of airspace safety assessment. Third, existing methods often rely on static data for modeling, while this invention, through spatiotemporal dynamic modeling, considers risk changes over different time periods, making airspace allocation more timely and adaptable. Furthermore, this invention combines eight-quadrant analysis and an improved Pareto ranking method to resolve multi-indicator conflict issues, optimizing airspace selection and decision-making mechanisms. Finally, this invention uses an improved Pareto ranking method for efficient airspace selection, generating high-resolution feasible airspace in large-scale urban airspace, significantly improving the safety and execution efficiency of UAV mission planning. Attached Figure Description

[0015] Figure 1 This represents the minimum number of hops in the ideal airspace of this invention.

[0016] Figure 2 This represents the minimum number of hops in the actual airspace of this invention.

[0017] Figure 3 This is a schematic diagram of daytime population distribution according to the present invention.

[0018] Figure 4 This is a schematic diagram of nighttime population distribution according to the present invention.

[0019] Figure 5 This is a schematic diagram of the spatial availability distribution according to Embodiment 1 of the present invention.

[0020] Figure 6 This is a schematic diagram of static risk distribution according to Embodiment 1 of the present invention.

[0021] Figure 7 This is a dynamic risk map for daytime periods according to Embodiment 1 of the present invention.

[0022] Figure 8 This is a dynamic risk map for the nighttime period in Embodiment 1 of the present invention.

[0023] Figure 9 This is a schematic diagram of the eight-quadrant analysis results during the daytime period in Embodiment 1 of the present invention.

[0024] Figure 10 This is a schematic diagram of the eight-quadrant analysis results during the nighttime period in Embodiment 1 of the present invention.

[0025] Figure 11 This is a three-dimensional map of the final usable airspace during the daytime period, according to Embodiment 1 of the present invention.

[0026] Figure 12 This is a three-dimensional map of the final usable airspace during the nighttime period, according to Embodiment 1 of the present invention. Detailed Implementation

[0027] The following detailed description of a specific embodiment of the present invention is provided in conjunction with the accompanying drawings. However, it should be understood that the scope of protection of the present invention is not limited to the specific embodiment.

[0028] This invention first quantifies airspace availability based on the shortest hop count measurement method. Then, it constructs two risk assessment models: a static risk model and a dynamic risk model. The static risk model considers factors such as LOS signal transmission risk, building collision risk, indirect fatality risk, and property damage risk, while the dynamic risk model accurately assesses direct fatality risk and social impact risk. Next, by combining the eight-quadrant analysis method with an improved Pareto ranking algorithm, urban airspace below 40 meters was modeled and evaluated. Finally, a 500×500×10 resolution three-dimensional feasible urban airspace map was generated.

[0029] This invention includes the following steps: S1: Airspace availability modeling involves three-dimensional discretization of the urban low-altitude area, dividing the study area into several 500×500×10 voxel grids with a layer spacing of 4m, corresponding to a total height of 40m. The spatial resolution of a single voxel is set to 10m×10m×4m for fine characterization of the urban low-altitude spatial structure. Based on actual Geographic Information System (GIS) building data, the height and location of buildings are mapped to corresponding voxel units, and obstacles are marked on voxels occupied by buildings. To quantify the navigability of urban low-altitude airspace in complex built environments, this invention employs a hop-count voxel-based airspace availability index.

[0030] The steps in S1 are as follows: S11: Construct a 5×5×5 (k=2) k-cuboid region centered on the evaluation voxel as the spatial analysis unit, and calculate the minimum hop count relationship between voxels; from voxels i Totozotoxin j The minimum number of hops is d unobs ( i , j ),like Figure 1 As shown; S12: In a real airspace containing static obstacles such as buildings, the minimum number of hops is denoted as... d obs ( i , j Due to the bypass effect, and satisfying d obs ( i , j )≥ d unobs ( i , j ),like Figure 2 As shown; S13: This invention constructs a spatial coverage index: (1); The closer the value is to 1, the weaker the impact of the obstacle on the navigation capability of the area. Furthermore, by averaging all effective voxels within the k-cuboid region, the overall voxel availability index is obtained: (2); in: Indicated by voxels The effective voxel set within the k-cuboid coverage area centered on the core. Effective voxels are voxel units not occupied by obstacles such as buildings. | represents the number of voxels in the set.

[0031] S2: Static risk model, used to assess risk factors in urban low-altitude airspace that do not change over time, mainly including line of sight (LOS) signal transmission risk, obstacle collision risk, indirect injury risk and property damage risk. The final static comprehensive risk value is obtained by scientifically allocating weights to the four types of risks using the entropy weight method.

[0032] S2 specifically includes: S21: Risks associated with line-of-sight signal transmission; Based on whether there is a Line-of-Sight (LOS) link between the antenna location and other grid cells, the communication risk is 0 when LOS exists; if the link is obstructed by a building, it is considered a communication interruption with a risk of 1. The mathematical model is as follows: (3); in; R 1 indicates a risk index for LOS signal transmission; Indicates "existence"; Indicates "does not exist"; LOS represents the unobstructed line-of-sight propagation path between the drone and the communication node; The communication risk is lowest when LOS exists and highest when LOS does not exist. S22: Risk of collision with obstacles; The closer the drone is to the obstacle, the higher the risk of collision; to avoid unlimited risk decay, a threshold is set. min If the distance between the drone and the target is greater than the threshold, it is considered to have no collision risk; if the distance is 0, the risk is recorded as 1. The obstacle collision risk is defined as follows: (4); in: R 2 indicates an obstacle collision risk index; dist_obs The minimum distance between the drone and city buildings; k GNSS For Global Navigation Satellite System (GNSS) accuracy adjustment factors; S23: Risk of indirect injury or death; Indirect injury risk refers to the potential injury to pedestrians caused by a drone crashing into a building. The risk model is as follows: (5); in: R 3 indicates an indirect injury risk indicator; P UCO This represents the probability of a building collision. N b_impact The number of buildings affected by the drone crash; N D P This is the average number of casualties coefficient; S24: Risk of property loss; Property damage risk refers to the potential losses caused by a drone colliding with various urban facilities due to system malfunctions during flight. Three loss scenarios are defined based on the type of ground object: (6); Property damage risk is calculated using the loss coefficient, collision kinetic energy, and system failure probability. (7); in: R 4 indicates a risk indicator of property loss; P crash This represents the failure probability of the unmanned aerial vehicle (UAV) system. E impact This refers to the actual collision energy. E max The maximum possible collision energy; S25: Static Risk Comprehensive Model; To avoid subjective weighting, the entropy weighting method is used to automatically calculate the weights. r i Its general formula is: (8); in: Indicates the first Weighting coefficients for static risk indicators; Indicates the first The information utility value of risk indicators; Indicates the first Information entropy of risk indicators; Indicates the first Individual unit in the first Normalized risk value under similar risk indicators; This represents the total number of voxel units involved in the risk calculation. It is a very small positive number, used to avoid numerical instability in logarithmic calculations; The final static risk model is: (9).

[0033] S3: Dynamic Risk Model, used to characterize the direct death risk and social impact risk of drones operating in urban low-altitude airspace due to the change in ground population distribution over time. It can reflect the pattern of human activity in the city at different times and realize the time-varying risk assessment of urban low-altitude airspace.

[0034] The specific steps of S3 are as follows: S31: In order to obtain the input parameters of the dynamic risk model, the urban population density is extrapolated in a secondary distribution based on land use type, so as to construct the ground population density model for different time periods. Considering the significant differences in daytime and nighttime activity patterns among urban residents; During the daytime, most people concentrate in industrial, commercial, and public activity areas, while a smaller number remain in residential areas, such as... Figure 3 As shown; At night, most people return to residential areas, while a minority remain in industrial, commercial, and public activity areas. For example... Figure 4 As shown; S32: Construction of dynamic risk indicators; Dynamic risks vary depending on population distribution at different times, including the direct risk of injury or death to pedestrians or vehicles caused by drone crashes. R 5. Social impact risks of drone noise on the ground population R The comprehensive assessment value of dynamic risk, consisting of 6 components, is defined as follows: (10); in: w 5. w 6 represents the weights of the two types of risks, satisfying... w 5+ w 6 = 1; The risk of direct death is due to the risk of injury or death to pedestrians. R fat_p Risk of vehicle injury or death R fat_v The weighted combination consists of: (11); pedestrian injury and death risk R fat_p Represented as: (12); in: P crash The probability of a drone crashing due to system malfunction; N hit p The number of pedestrians within the area affected by the fall; R f P The fatality rate of pedestrians after being hit; Vehicle injury risk R fat_v Represented as: (13); in: N hit v Vehicles within the impact area of ​​the fall; R f v The fatality rate of pedestrians after being hit; Social impact risk is used to measure the disturbance caused by drone noise in densely populated areas on the ground. Its calculation formula is as follows: (14); in: L hFor reference noise intensity; ω This is the conversion factor from noise to impact level; H This refers to the flight altitude of the drone. d noise The horizontal distance between the drone and the center of the affected grid.

[0035] S4: Eight-quadrant analysis and an improved Pareto ranking method, based on airspace availability, to achieve rapid classification of urban low-altitude airspace. α Static risks R static ) and dynamic risk ( R dynamic A four-quadrant classification system combining three indicators is constructed, with each indicator divided into two levels: high (H) and low (L), resulting in eight combinations: LLL, LLH, LHL, LHH, HLL, HLH, HHL, and HHH. Among them, the LLH category (low static risk, low dynamic risk, and high availability) represents the preferred airspace region. Based on quadrant classification, to further improve the accuracy and efficiency of spatial domain screening, an improved Pareto ranking method is introduced. By constructing a linear objective function to solve the three types of indicators in a weighted manner, a comprehensive ranking of voxel units is achieved.

[0036] In S4, the objective function of the improved Pareto sorting method is: (15); in: f 1、 f 2 represents the normalized value of static risk and dynamic risk; α This is an indicator of airspace availability. w 1, w 2, w 3 is the weighting coefficient, which satisfies .

[0037] Example 1: To verify the effectiveness of the constructed airspace availability index, multi-risk assessment model, and four-quadrant and improved Pareto ranking method, this section conducts an experimental analysis based on a real urban area. The experiment selects a 500×500×10 three-dimensional voxelized urban airspace with a layer height of 4m and a total height of 40m. Through a comprehensive comparison of airspace availability, static risk, dynamic risk, and feasible airspace, the rationality and practicality of the proposed method are verified.

[0038] S1: Spatial availability distribution; Based on a voxel-based usability calculation method, the airspace usability at 10 altitude levels was evaluated. The results show that the lower levels (4-12 m) are affected by building obstruction, resulting in lower usability and significant spatial discontinuity. Usability gradually increases with altitude, approaching 1 at higher levels (28-40 m), indicating that the high-altitude airspace is almost unobstructed and has good connectivity. Figure 5 As shown.

[0039] S2: Static risk distribution; Static risks include LOS signal transmission risk, obstacle collision risk, indirect injury risk, and property damage risk. Experimental results show that static risks are significantly higher in lower-level areas, mainly due to high building density and easy LOS obstruction; risks decrease significantly in higher-level areas due to the relatively open environment; however, some high-risk areas still exist locally due to obstacle / communication risks caused by tall buildings, as follows: Figure 6 As shown.

[0040] S3: Dynamic risk distribution; Dynamic risk maps are generated based on population activity patterns during daytime and nighttime.

[0041] The risk is significantly higher during the day than at night because people are mainly concentrated in industrial areas, office areas, and public activity areas; the dynamic risk decreases with increasing altitude; the risk at night is mainly concentrated in residential areas, with a smaller scope and lower intensity. The dynamic risk map for daytime periods is shown below. Figure 7 As shown, the dynamic risk map for the nighttime period is as follows: Figure 8 As shown.

[0042] S4: S41; Results of Eight-Quadrant Analysis Based on thresholds set for availability, static risk, and dynamic risk, all voxels were divided into 8 categories. A final 3D map of the available airspace was generated. Results show that the LLH (low static risk, low dynamic risk, high availability) region is mainly distributed in the mid-to-high altitudes; due to the simultaneously high static and dynamic risks in the lower atmosphere, the LLH region is extremely rare.

[0043] S42: Final Available 3D Spatial Map Based on the eight-quadrant screening results, an improved Pareto objective function was further used to comprehensively rank the remaining voxels, selecting the top 70% as the final usable airspace. Experimental results show that the final usable airspace exhibits good continuity at altitudes of 20–40m; although some local low-level usable airspace exists, it is small and scattered, unsuitable for large-scale UAV route layout; the final results are significantly better than the spatial quality based solely on a single indicator screening.

[0044] The above-disclosed embodiments are merely specific examples of the present invention. However, the present invention is not limited thereto, and any variations that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A method for evaluating urban airspace based on availability quantification and multi-risk assessment, characterized in that, Includes the following steps: S1: Airspace availability modeling, which performs three-dimensional discretization on the urban low-altitude area and divides the study area into several voxel grids for fine characterization of the urban low-altitude spatial structure. Based on actual geographic information system building data, the height and location of buildings are mapped to corresponding voxel units, and obstacles are marked on the voxels occupied by buildings. S2: Static risk model, used to assess risk factors in urban low-altitude airspace that do not change over time, mainly including line-of-sight signal transmission risk, obstacle collision risk, indirect injury risk and property damage risk. The final static comprehensive risk value is obtained by scientifically allocating weights to the four types of risks using the entropy weight method. S3: Dynamic risk model, which can reflect the patterns of human activity in the city at different times and realize time-varying risk assessment of urban low-altitude airspace; S4: Eight-quadrant analysis and improved Pareto ranking method. Based on airspace availability, static risk and dynamic risk, a four-quadrant classification system is constructed that combines three indicators. The three indicators are divided into high and low levels, forming a total of eight combinations. Based on quadrant classification, to further improve the accuracy and efficiency of spatial domain screening, an improved Pareto ranking method is introduced. By constructing a linear objective function to solve the three types of indicators in a weighted manner, a comprehensive ranking of voxel units is achieved.

2. The urban airspace evaluation method based on availability quantification and multi-risk assessment according to claim 1, characterized in that: The steps in S1 are as follows: S11: Construct a 5×5×5 (k=2) k-cuboid region centered on the evaluation voxel as the spatial analysis unit, and calculate the minimum hop count relationship between voxels; from voxels i Totozotoxin j The minimum number of hops is d unobs ( i , j ); S12: In a real airspace containing static obstacles such as buildings, the minimum number of hops is denoted as... d obs ( i , j Due to the bypass effect, and satisfying d obs ( i , j )≥ d unobs ( i , j ); S13: This invention constructs a spatial coverage index: （1）； The overall voxel availability index is obtained by averaging all available voxels within the k-cuboid region. （2）； in: Indicated by voxels The effective voxel set within the k-cuboid coverage area centered on the core. Effective voxels are voxel units not occupied by building obstacles. | represents the number of voxels in the set.

3. The urban airspace evaluation method based on availability quantification and multi-risk assessment according to claim 2, characterized in that: S2 specifically includes: S21: Risks associated with line-of-sight signal transmission; Based on whether there is a Line-of-Sight (LOS) link between the antenna location and other grid cells, the communication risk is 0 when LOS exists; if the link is obstructed by a building, it is considered a communication interruption with a risk of 1. The mathematical model is as follows: （3）； in; R 1 indicates the LOS signal transmission risk index; Indicates "existence"; Indicates "does not exist"; LOS represents the unobstructed line-of-sight propagation path between the drone and the communication node; The communication risk is lowest when LOS exists and highest when LOS does not exist. S22: Risk of collision with obstacles; The closer the drone is to the obstacle, the higher the risk of collision; to avoid unlimited risk decay, a threshold is set. min If the distance between the drone and the target is greater than the threshold, it is considered to have no collision risk; if the distance is 0, the risk is recorded as 1. The obstacle collision risk is defined as follows: （4）； in: R 2 indicates an obstacle collision risk index; dist_obs The minimum distance between the drone and city buildings; k GNSS For global satellite navigation system accuracy adjustment factors; S23: Risk of indirect injury or death; The risk model is as follows: （5）； in: R 3 indicates an indirect injury risk indicator; P UCO This represents the probability of a building collision. N b_impact The number of buildings affected by the drone crash; N D P The average number of casualties is the coefficient. S24: Risk of property loss; Based on the type of ground object, three loss scenarios are defined: （6）； Property damage risk is calculated using the loss coefficient, collision kinetic energy, and system failure probability. （7）； in: R 4 indicates a risk indicator of property loss; P crash This represents the failure probability of the unmanned aerial vehicle (UAV) system. E impact This refers to the actual collision energy. E max The maximum possible collision energy; S25: Static Risk Comprehensive Model; The weights are automatically calculated using the entropy weight method. r i Its general formula is: （8）； in: Indicates the first Weighting coefficients for static risk indicators; Indicates the first The information utility value of risk indicators; Indicates the first Information entropy of risk indicators; Indicates the first Individual unit in the first Normalized risk value under similar risk indicators; This represents the total number of voxel units involved in the risk calculation. It is a very small positive number, used to avoid numerical instability in logarithmic calculations; The final static risk model is: （9）。 4. The urban airspace evaluation method based on availability quantification and multi-risk assessment according to claim 2, characterized in that: The specific steps of S3 are as follows: S31: In order to obtain the input parameters of the dynamic risk model, the urban population density is extrapolated in a secondary distribution based on land use type, so as to construct the ground population density model for different time periods. S32: Construction of dynamic risk indicators; Dynamic risks vary depending on population distribution at different times, including the direct risk of injury or death to pedestrians or vehicles caused by drone crashes. R 5. Social impact risks of drone noise on the ground population R The comprehensive assessment value of dynamic risk, consisting of 6 components, is defined as follows: （10）； in: w 5. w 6 represents the weights of the two types of risks, satisfying... w 5+ w 6 = 1; The risk of direct death is due to the risk of injury or death to pedestrians. R fat_p Risk of vehicle injury or death R fat_v The weighted combination consists of: （11）； pedestrian injury and death risk R fat_p Represented as: （12）； in: P crash The probability of a drone crashing due to system malfunction; N hit p The number of pedestrians within the area affected by the fall; R f P The fatality rate of pedestrians after being hit; Vehicle injury risk R fat_v Represented as: （13）； in: N hit v Vehicles within the impact area of ​​the fall; R f v The fatality rate of pedestrians after being hit; Social impact risk is used to measure the disturbance caused by drone noise in densely populated areas on the ground. Its calculation formula is as follows: （14）； in: L h For reference noise intensity; ω This is the conversion factor from noise to impact level; H This refers to the flight altitude of the drone. d noise The horizontal distance between the drone and the center of the affected grid.

5. The urban airspace evaluation method based on availability quantification and multi-risk assessment according to claim 4, characterized in that: In S4, the objective function of the improved Pareto sorting method is: （15）； in: f 1、 f 2 represents the normalized value of static risk and dynamic risk; α This is an indicator of airspace availability. w 1, w 2, w 3 represents the weighting coefficient.

Images