An anisotropic risk assessment method for unmanned aerial vehicle operation in urban low-altitude scenarios

Abstract

The application provides a kind of city low-altitude scene under the anisotropic risk assessment method of unmanned aerial vehicle operation, comprising: obtaining unmanned aerial vehicle parameters and flight environment characteristic data, and constructing initial ground risk map;Based on the initial ground risk map, horizontal ground risk field and vertical ground risk field are respectively constructed;According to the horizontal ground risk field and the vertical ground risk field, anisotropic ground risk field is constructed;The anisotropic ground risk field is converted into ground risk vector field by using the anisotropic ground risk field;Real-time risk calculation in any flight direction is realized by the ground risk vector field.The application accurately quantifies the risk difference of different flight directions, provides real-time and accurate directional risk basis for unmanned aerial vehicle path planning, and significantly improves the safety and airspace utilization efficiency of city low-altitude operation.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios. Background Technology

[0002] With the rapid development of drone technology, its applications in urban low-altitude logistics, inspection, and emergency rescue are becoming increasingly widespread. However, the risk of ground casualties and facility damage caused by drone crashes has become a key bottleneck restricting its large-scale application. Traditional risk assessment methods often use scalar risk maps, which only quantify the magnitude of risk while ignoring directional sensitivity. This leads to problems such as wasted airspace or excessive conservatism in path planning, for example, a uniform safety margin cannot adapt to the risk differences in different directions.

[0003] In existing technologies, risk assessment mainly relies on static parameters such as population density and obstacle distribution to generate a scalar risk field, displaying the risk gradient through contour lines. However, it does not consider the impact of the drone's flight direction on the risk. For example, the crash risk when flying parallel to a building complex may be significantly lower than that in the vertical direction, but the scalar risk map cannot distinguish such differences, leading to an imbalance between the safety and economy of path planning.

[0004] Therefore, there is an urgent need for an assessment method that can quantify the directional characteristics of risks. By constructing a directional vector risk field, the risk of any flight direction can be calculated quickly, providing accurate directional references for UAV path planning. Summary of the Invention

[0005] This invention provides a method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios, in order to overcome the shortcomings of traditional risk assessment in the prior art that ignores directional differences.

[0006] In a first aspect, the present invention provides a method for assessing the anisotropic risk of unmanned aerial vehicle (UAV) operation in urban low-altitude scenarios, comprising: Acquire drone parameters and flight environment characteristics data to construct an initial ground risk map; Based on the initial ground risk map, a horizontal ground risk field and a vertical ground risk field are constructed respectively; An anisotropic ground risk field is constructed based on the horizontal ground risk field and the vertical ground risk field; The anisotropic ground risk field is converted into a ground risk vector field. The ground risk vector field enables real-time risk calculation for any flight direction.

[0007] According to the present invention, a method for assessing the anisotropic risk of unmanned aerial vehicle (UAV) operation in urban low-altitude scenarios is provided, which acquires UAV parameters and flight environment characteristic data, and constructs an initial ground risk map, including: The low-altitude ground area of ​​the city to be assessed is discretized in two dimensions. The preset airspace interval is used as the grid unit scale. Multiple cell matrices are formed based on ground risk factors. Each cell corresponds to a unique geographic reference unit. The matrix elements in the cell matrix are the risk values ​​of the cell's location. Linearly convert the discrete coordinates of the matrix elements into geographic location coordinates; Based on the aforementioned geographical coordinates, and using a weighted calculation based on population density and building asset value, the risk level of casualties among ground personnel is quantified to obtain the population density layer. Based on the geographical coordinates, the ability of surface shelters to protect people on the ground is quantified to obtain the shelter layer; Based on the geographical coordinates, the height distribution of obstacles is quantified to obtain the obstacle layer; Based on the aforementioned geographical coordinates, the distribution of no-fly zones is quantified to obtain no-fly layers; Based on the drone's cruising altitude, the population density layer, the shielding layer, the obstacle layer, and the no-fly zone are merged to generate an initial ground risk map that includes the test flight zone, the restricted flight zone, and the no-fly zone.

[0008] According to the present invention, a method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios is provided, which constructs a horizontal ground risk field, including: Determine the horizontal flight direction angle of the drone and quantify the collision risk when the drone flies horizontally along the edge of the no-fly zone and restricted flight zone; Using UAV positioning error, size and mass, obstacle positioning error, and wind field influence as core indicators, the coordinates of the target obstacle and the UAV position are obtained. The positioning error is determined to follow a normal distribution. The distance error and standard deviation are calculated from the target obstacle position coordinates and the UAV position coordinates. A simplified standard deviation is obtained based on the standard deviation of the obstacle positioning error and the standard deviation of the UAV positioning error. By combining the collision probability, wind field correction factor, simplified standard deviation, maximum radius of the UAV, and equivalent radius of the obstacle, the horizontal safety interval is obtained; The no-fly zone and restricted-fly zone are horizontally dilated, with the dilation scale based on the horizontal safety interval. Multiple risk levels are divided, and a monotonically decreasing scale function of the horizontal distance from the current point to the nearest no-fly or restricted-fly boundary is solved to obtain the horizontal ground risk field.

[0009] According to the present invention, a method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios is provided, which constructs a vertical ground risk field, including: Determine the vertical flight direction angle of the drone and quantify the collision risk when the drone flies vertically along the edge of the no-fly zone and restricted flight zone; A dual-objective optimization model constrained by multiple conditions is constructed based on the drone boundary crossing conflict rate, buffer airspace ratio, cruising speed, size and mass, distance and wind field influence. The vertical safety interval is obtained by solving the dual-objective optimization model. The dual-objective optimization model includes the optimization objective of minimum boundary collision rate and the optimization objective of minimum buffer space ratio. The multiple constraints include deceleration distance condition, collision judgment condition, positioning error probability condition, velocity and acceleration constraint condition, and safety interval value constraint condition. The no-fly zone and restricted-fly zone are data-dilated in terms of vertical distance. The dilation scale is based on the vertical safety interval and multiple risk levels are divided. The monotonically decreasing scale function of the vertical distance from the current point to the nearest no-fly or restricted-fly boundary is solved to obtain the vertical ground risk field.

[0010] According to the present invention, a method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios is provided, which constructs an anisotropic ground risk field based on the horizontal ground risk field and the vertical ground risk field, including: In the local coordinate system, the horizontal and vertical directions are determined to be mutually orthogonal principal axes, wherein the horizontal direction is parallel to the boundary, the vertical direction is perpendicular to the boundary, and the horizontal ground risk field and the vertical ground risk field are cosine complementary. Determine the adjustable shape parameters, and obtain the horizontal angle weight and vertical angle weight from the shape parameters, the angle between the UAV flight direction and the horizontal direction, the horizontal ground risk field and the vertical ground risk field; The directional coupling correction term is obtained from the angle between the UAV's flight direction and the horizontal direction, the coupling coefficient, the horizontal ground risk field, and the vertical ground risk field; The anisotropic ground risk field is constructed by combining the horizontal angle weight, the vertical angle weight, the directional coupling correction term, the horizontal ground risk field, and the vertical ground risk field. The anisotropic ground risk field is constrained by boundary condition constraints and diagonal fly-through constraints.

[0011] According to the present invention, a method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios is provided, which utilizes the anisotropic ground risk field to convert it into a ground risk vector field, including: The maximum upward direction of the anisotropic ground risk field in the plane is obtained by using gradients, wherein the gradients in the x and y directions are obtained by central difference on the grid. Determine a noise threshold, and constrain the maximum upward direction with the noise threshold to obtain a unit vector of the direction with the fastest wind direction growth; The ground risk vector field is obtained by fusing the anisotropic ground risk field and the unit vector of the direction of fastest wind growth.

[0012] According to the present invention, a method for assessing the anisotropic risk of unmanned aerial vehicle (UAV) operation in urban low-altitude scenarios is provided, which realizes real-time risk calculation for any flight direction using the ground risk vector field, including: The ground risk vector field is decomposed into eigenvalues ​​to obtain a first orthogonal eigenvector and a first eigenvalue, as well as a second orthogonal eigenvector and a second eigenvalue. The first orthogonal eigenvector corresponds to the direction of maximum risk, and the second orthogonal eigenvector corresponds to the direction of minimum risk. Calculate the cosine and sine values ​​of the angle between the UAV's flight direction and the horizontal direction to obtain the unit vectors of the UAV's flight direction and the horizontal direction; Based on the first orthogonal eigenvector, the first eigenvalue, the second orthogonal eigenvector, the second eigenvalue, and the unit vectors of the UAV's flight direction and horizontal direction, a fast interpolation calculation is performed to obtain the risk value of the UAV in any flight direction.

[0013] Secondly, the present invention also provides a system for assessing the anisotropic risk of unmanned aerial vehicle (UAV) operation in urban low-altitude scenarios, comprising: The acquisition module is used to acquire UAV parameters and flight environment characteristic data to construct an initial ground risk map; The first construction module is used to construct a horizontal ground risk field and a vertical ground risk field based on the initial ground risk map. The second construction module is used to construct an anisotropic ground risk field based on the horizontal ground risk field and the vertical ground risk field. The conversion module is used to convert the anisotropic ground risk field into a ground risk vector field. The calculation module is used to perform real-time risk calculations for any flight direction based on the ground risk vector field.

[0014] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the anisotropic risk assessment method for UAV operation in urban low-altitude scenarios as described above.

[0015] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios as described above.

[0016] This invention provides a method for assessing the anisotropic risk of UAV operations in urban low-altitude scenarios. It constructs an initial ground risk map by combining a multi-dimensional ground attribute map, then quantifies the collision risk in the horizontal and vertical flight directions to build horizontal and vertical ground risk fields. By establishing a coupling mathematical function between these two fields and fusing flight direction angle features, an anisotropic ground risk field is generated. This field is further converted into a direction-sensitive ground risk vector field, and finally, based on orthogonal decomposition of eigenvectors, rapid risk interpolation for any flight direction is achieved. This invention accurately quantifies the risk differences between different flight directions, providing real-time and precise directional risk information for UAV path planning, significantly improving the safety of urban low-altitude operations and airspace utilization efficiency. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is one of the flowcharts illustrating the anisotropic risk assessment method for UAV operation in urban low-altitude scenarios provided by the present invention. Figure 2 This is the second flowchart of the method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios provided by the present invention. Figure 3 This is a schematic diagram of the risk diagram representation method provided by the present invention; Figure 4 This is a schematic diagram of the initial ground value provided by the present invention; Figure 5 This is a schematic diagram of the horizontal and vertical ground risk fields provided by the present invention; Figure 6 This is a schematic diagram of the diagonal fly-over constraint provided by the present invention; Figure 7 This is a schematic diagram of the anisotropic risk field and its risk value provided by the present invention; Figure 8 This is a diagram showing the ground risk field results generated in different flight directions, provided by the present invention. Figure 9 This is a schematic diagram of the structure of the anisotropic risk assessment system for UAV operation in urban low-altitude scenarios provided by the present invention; Figure 10 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0020] Figure 1 This is one of the flowcharts illustrating the anisotropic risk assessment method for UAV operation in urban low-altitude scenarios provided in this embodiment of the invention, such as... Figure 1 As shown, it includes: Step 100: Acquire UAV parameters and flight environment characteristic data to construct an initial ground risk map; Step 200: Based on the initial ground risk map, construct the horizontal ground risk field and the vertical ground risk field respectively; Step 300: Construct an anisotropic ground risk field based on the horizontal ground risk field and the vertical ground risk field; Step 400: Convert the anisotropic ground risk field into a ground risk vector field; Step 500: Real-time risk calculation for any flight direction is achieved using the ground risk vector field.

[0021] The embodiments of the present invention include the following steps: S1: Constructing an initial ground risk map by combining a multi-dimensional ground attribute map: The area to be evaluated is discretized in two dimensions, using a preset airspace interval as the grid unit scale. Ignoring UAV crash risk, the space is divided into grids based solely on ground risk factors. Each grid is assigned ground risk attributes including population density, obstruction level, obstacle density, and no-fly zones. An initial two-dimensional ground risk grid map with test flight zones, restricted flight zones, and no-fly zones is then generated according to a hierarchical quantification system. ; S2: Constructing a horizontal ground risk field: Taking no-fly zones and restricted flight zones as the main research subjects, quantifying the collision risk level of UAVs when flying horizontally along the edges of no-fly zones and restricted flight zones, i.e., the UAV flight direction First, using the drone's positioning error, drone size and mass, positioning error of obstacles, and wind field influence as the main metrics, the minimum horizontal safe distance for the drone at the current cruising altitude is calculated. S Then, with a horizontal safety interval S The no-fly zones and restricted-fly zones are mathematically expanded into five levels, with different risk levels assigned based on their distance from the no-fly zones and restricted-fly zones, forming a horizontal ground risk field. ; S3: Constructing a Vertical Ground Risk Field: Taking no-fly zones and restricted flight zones as the main research subjects, quantifying the collision risk level when UAVs fly perpendicular to the edges of no-fly zones and restricted flight zones, i.e., the UAV flight direction. First, the impact of drone boundary crossing conflict rate and drone buffer airspace occupancy on the safe buffer distance for direct flight is considered. Second, the drone's cruising speed, mass and size, distance between the drone and obstacles, and wind field influence are used as the main metrics to optimize the safe buffer distance for direct flight, obtaining the minimum vertical safe separation. D Finally, with vertical safety intervals. D The no-fly zones and restricted-fly zones are mathematically expanded into five levels, with different risk levels assigned based on their distance from the no-fly zones and restricted-fly zones, forming a vertical ground risk field. ; S4: Constructing an anisotropic ground risk field: Since the horizontal ground risk field records the risk level of the minimum safe separation in the UAV's flight direction, while the vertical ground risk field records the risk level of the maximum safe separation in the UAV's flight direction, it is natural that for any UAV flight direction... Specifically, the risk value can be calculated based on the angle in any UAV flight direction. When horizontal and vertical risks are considered as orthogonal principal directions, and interpolated using an angle weighting function, the risk value in any flight direction can be calculated. All of these yielded continuous, differentiable, and physically meaningful comprehensive risk values, ultimately resulting in an anisotropic ground risk field. ; S5: Using the anisotropic ground risk field obtained in step S4, convert it into a direction-sensitive ground risk vector field. This adds a unit vector in the fastest direction to the ground risk. This vector field reflects the effective intensity of the ground risk in a specific flight direction and is a component of the final "anisotropic risk field." S6: Fast interpolation of risk for arbitrary flight directions: When the drone's flight direction is... At that time, the risk value is calculated by orthogonal decomposition of eigenvectors, realizing real-time risk calculation for any flight direction, which is used for path planning and flight safety assessment.

[0022] Specifically, such as Figure 2 As shown, it includes: Step 1: Constructing the Initial Ground Risk Map: The low-altitude area of ​​the target city is divided into two-dimensional raster layers. These layers are then combined with multiple ground attribute layers, including population density, obstructions, obstacle height, and no-fly zones, to create an initial ground risk distribution map encompassing three main categories: test flight zones, restricted flight zones, and no-fly zones. Details are as follows: S1: Construct an initial ground risk map by combining the multi-dimensional ground attribute map: S1.1 Two-dimensional rasterization: The low-altitude ground area of ​​the city to be evaluated is discretized in two dimensions, with a preset spatial interval. l The grid cell size (5m-15m) is used, without considering the risk of UAV crashes, and is formed solely based on ground risk factors. M × N cell matrix R Each cell corresponds to a unique geographic reference unit, matrix elements R ( i , j The risk value for the cell's location is ( ) ),like Figure 3 As shown; S1.2 Coordinate Transformation Relationship: Discrete Coordinates ( i , j ) and geographical location x , y The conversion formula for ) is:

[0023] S1.3 Construction of Multidimensional Ground Attribute Layers: Generate four types of core ground attribute layers, all of which are location-based two-dimensional maps: (1) Population density layer : Quantify the risk level of casualties among ground personnel based on a weighted calculation of population density and building asset value; (2) Shielding layer : Quantify the ability of surface shelters to protect people on the ground (value ranges from 0 to 1, with larger values ​​indicating stronger protection). (3) Barrier layer : Quantize obstacle height distribution (0-1 binarization, 1 indicates the presence of an obstacle); (4) No-fly zone : Quantize the distribution of no-fly zones (0-1 binary layer, 0 represents no-fly zone, 1 represents flyable zone); S1.4 Initial Ground Risk Map Fusion: Merge the above layers according to the following rules to generate an initial two-dimensional ground risk raster map containing the test flight area, restricted flight area, and no-fly zone. R ( x , y ):

[0024] in h This represents the drone's cruising altitude. 0 indicates that flight is prohibited due to no-fly zones or obstacles exceeding the cruising altitude. Figure 4 As shown.

[0025] Step 2: Constructing a Horizontal Risk Field: For scenarios where drones fly horizontally along obstacle boundaries or restricted / no-fly zones, considering factors such as drone and obstacle positioning errors, drone size, and wind disturbances, the minimum horizontal safety interval is calculated. This interval is then used as a scale to perform mathematical morphological expansion of the restricted / no-fly zone, distinguishing risk levels at different distances to generate a horizontal risk field. Details are as follows: S2: Constructing a horizontal ground risk field : S2.1 Definition of Horizontal Flight Scenario: Quantify the collision risk (flight direction) of a drone flying horizontally along the edge of a no-fly zone or restricted flight zone. ); S2.2 Horizontal safety interval S Calculation: Using UAV positioning error, size and mass, obstacle positioning error, and wind field influence as core indicators, for target obstacles ( ) and drone location information ( Assuming that its positioning error follows a normal distribution, the distance error and standard deviation are derived as follows:

[0026]

[0027] make (Standard deviation of obstacle localization error) (Standard deviation of UAV positioning error), simplified to: .

[0028] Introducing the probability of collision occurrence ( (for positioning deviation), combined with wind field correction factor (Tailwind: 1.2, Headwind: 0.8, Crosswind: 1.0), Final horizontal safety separation:

[0029] in The maximum radius of the drone Let be the equivalent radius of the obstacle.

[0030] S2.3 Risk Field Expansion and Classification: The no-fly zone / restricted-fly zone is mathematically expanded (morphological dilation) according to the horizontal distance, with the expansion scale based on the horizontal safety interval. S Based on this standard, it is divided into five levels: S, 2S, 3S, 4S 5S Corresponding risk level arrive The mathematical expression is:

[0031] in The horizontal distance from the current point to the nearest no-fly / restricted-fly boundary. For a monotonically decreasing scale function, such as Figure 5 As shown.

[0032] S3: Constructing a Vertical Ground Risk Field : S3.1 Vertical Flight Scenario Definition: Quantifying the collision risk (flight direction) when a drone flies perpendicular to the edge of a no-fly zone or restricted flight zone. ); S3.2 Vertical safety interval D Optimization calculation: Considering the drone's boundary crossing conflict rate, buffer airspace occupancy, cruising speed, its own mass and size, distance, and wind field influence, a dual-objective optimization model is constructed: Optimization objective 1 (minimum boundary violation rate);

[0033] Optimization objective 2 (minimum buffer space ratio);

[0034] Constraints: (Deceleration distance condition) (Conditions for determining conflict) (Positioning error probability condition) (Velocity and acceleration constraints) (Constraints on the value of the safety interval) in For conflict indicator functions ( (Take 1 if it is true, otherwise take 0). Here, a, b, and c represent the conflict event weights (1 for conflicts), and a, b, and c are the three-dimensional volume parameters of the obstacles. For the size of the drone protection zone, , The mean and standard deviation of the UAV positioning error are given; a wind field correction factor is introduced. (1.5 when perpendicular to the building wall, 1.8 when there is significant wind turbulence), final vertical safety clearance ( (Optimal solution for dual objectives) S3.3 Risk Field Expansion and Classification: The no-fly zone / restricted-fly zone is mathematically expanded (morphological dilation) according to the horizontal distance, with the expansion scale based on the vertical safety interval. Based on this standard, it is divided into five levels: Corresponding risk level arrive Mathematical expression:

[0035] in The vertical distance from the current point to the nearest no-fly / restricted-fly boundary. For a monotonically decreasing scale function, such as Figure 5 As shown.

[0036] Step 4: Construct an anisotropic ground risk field: Treating horizontal and vertical risks as orthogonal principal directions, and using an angle weighting function for interpolation coupling, a continuous, differentiable, and physically meaningful comprehensive risk value is obtained in any flight direction, thus constructing a direction-sensitive anisotropic risk field. Details are as follows: S4: Construct an anisotropic ground risk field : S4.1 Risk Coupling Assumption: In the local coordinate system, the "horizontal direction" (parallel to the boundary) and the "vertical direction" (perpendicular to the boundary) are considered as mutually orthogonal principal axes; when The risk should approach ,when The risk should approach The two are complementary cosines.

[0037] S4.2 Mathematical Function Construction: First, use angle weights , Perform angle interpolation and use adjustable shape parameters. p To control the "steepness".

[0038]

[0039]

[0040] in p ≥1 is a shape parameter. This construction guarantees... , and when hour ,when hour Construct an anisotropic risk calculation model:

[0041] in: The angle between the drone's flight direction and the horizontal direction (0°≤ ≤90°); For the directional coupling correction term, consider the nonlinear superposition effect of horizontal and vertical risks:

[0042] The coupling coefficient (values ​​range from 0.1 to 0.3, with 0.3 for densely populated urban areas and 0.1 for suburban areas). S4.3 Boundary Condition Constraints: When hour, ;when hour, When 0° < When the angle is less than 90°, the risk value changes smoothly with the angle.

[0043] S4.5 Diagonal Flight Constraint: Since the diagonal point is no longer within the parallel flight protection zone, only flying completely along the diagonal will result in the greatest vertical flight risk at the diagonal point. Therefore, only the vertical flight risk is considered at the diagonal point, and the horizontal flight risk is no longer considered. Figure 6 As shown.

[0044] Step 5: Construct the risk vector field: Perform gradient calculations on anisotropic risks to obtain the unit vector of the direction of fastest risk growth under ground constraints. Transform the risk scalar field into a risk vector field, enabling the risk to express both direction and intensity. Details are as follows: S5: Construct an anisotropic risk vector field : S5.1 Risk gradient calculation: For anisotropic ground risk fields It can be viewed as a parameterization of a two-dimensional scalar field in terms of angle, such as Figure 7 As shown, its maximum ascent direction in the plane can be obtained using the gradient:

[0045] The gradient is implemented in numerical calculations using central difference (on a grid):

[0046] S5.2 The unit vector in the direction of fastest risk growth is defined as:

[0047] in Set a threshold to avoid noise amplification. If you want the direction to be sensitive to the flight direction, you can set... In Take the local average value within the current aircraft direction or direction window.

[0048] S5.3 Vector Field Construction: Fusing risk values ​​and unit vectors to generate a ground risk vector field:

[0049] This vector field reflects the effective intensity and directional characteristics of ground risks under a specific flight direction.

[0050] Step 6: Achieve rapid risk calculation in any direction: Real-time directional interpolation is achieved by utilizing the orthogonal decomposition of the eigenvectors of the risk vector field. This allows the UAV to directly calculate risk in any flight direction without omnidirectional sampling, providing real-time support for UAV online path planning, attitude decision-making, and flight safety assessment. Details are as follows: S6: Rapid interpolation of risks in any flight direction: S6.1 Eigenvector Orthogonal Decomposition: For the ground risk vector field Eigenvalue decomposition yields two orthogonal eigenvectors. , and corresponding eigenvalues , ( ≥ ),in The direction with the greatest corresponding risk The direction with the lowest corresponding risk; S6.2 Flight Direction Unit Vector: UAV in any flight direction Unit vector:

[0051] S6.3 Fast Interpolation Calculation: Risk values ​​for any flight direction are quickly calculated using orthogonal decomposition of eigenvectors. The formula is as follows: It achieves real-time calculation of risk in any flight direction without requiring omnidirectional sampling, supporting path planning and flight safety assessment, such as... Figure 8 As shown.

[0052] The anisotropic risk assessment system for UAV operation in urban low-altitude scenarios provided by this invention is described below. The anisotropic risk assessment system for UAV operation in urban low-altitude scenarios described below can be referred to in correspondence with the anisotropic risk assessment method for UAV operation in urban low-altitude scenarios described above.

[0053] Figure 9 This is a schematic diagram of the structure of the anisotropic risk assessment system for UAV operation in urban low-altitude scenarios provided in an embodiment of the present invention, as shown below. Figure 9 As shown, it includes: an acquisition module 91, a first construction module 92, a second construction module 93, a conversion module 94, and a calculation module 95, wherein: The acquisition module 91 is used to acquire UAV parameters and flight environment characteristic data to construct an initial ground risk map; the first construction module 92 is used to construct a horizontal ground risk field and a vertical ground risk field based on the initial ground risk map; the second construction module 93 is used to construct an anisotropic ground risk field based on the horizontal ground risk field and the vertical ground risk field; the conversion module 94 is used to convert the anisotropic ground risk field into a ground risk vector field; and the calculation module 95 is used to perform real-time risk calculation for any flight direction using the ground risk vector field.

[0054] Figure 10 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 10 As shown, the electronic device may include a processor 1010, a communications interface 1020, a memory 1030, and a communication bus 1040, wherein the processor 1010, communications interface 1020, and memory 1030 communicate with each other via the communication bus 1040. The processor 1010 can call logical instructions in the memory 1030 to execute an anisotropic risk assessment method for UAV operation in urban low-altitude scenarios. This method includes: acquiring UAV parameters and flight environment characteristic data to construct an initial ground risk map; constructing a horizontal ground risk field and a vertical ground risk field based on the initial ground risk map; constructing an anisotropic ground risk field based on the horizontal and vertical ground risk fields; converting the anisotropic ground risk field into a ground risk vector field; and using the ground risk vector field to perform real-time risk calculation for any flight direction.

[0055] Furthermore, the logical instructions in the aforementioned memory 1030 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0056] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the anisotropic risk assessment method for UAV operation in urban low-altitude scenarios provided by the methods described above. The method includes: acquiring UAV parameters and flight environment characteristic data, and constructing an initial ground risk map; constructing a horizontal ground risk field and a vertical ground risk field based on the initial ground risk map; constructing an anisotropic ground risk field based on the horizontal and vertical ground risk fields; converting the anisotropic ground risk field into a ground risk vector field; and performing real-time risk calculation for any flight direction using the ground risk vector field.

[0057] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0058] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0059] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios, characterized in that, include: Acquire drone parameters and flight environment characteristics data to construct an initial ground risk map; Based on the initial ground risk map, a horizontal ground risk field and a vertical ground risk field are constructed respectively; An anisotropic ground risk field is constructed based on the horizontal ground risk field and the vertical ground risk field; The anisotropic ground risk field is converted into a ground risk vector field. The ground risk vector field enables real-time risk calculation for any flight direction.

2. The method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios according to claim 1, characterized in that, Acquire UAV parameters and flight environment characteristics data to construct an initial ground risk map, including: The low-altitude ground area of ​​the city to be assessed is discretized in two dimensions. The preset airspace interval is used as the grid unit scale. Multiple cell matrices are formed based on ground risk factors. Each cell corresponds to a unique geographic reference unit. The matrix elements in the cell matrix are the risk values ​​of the cell's location. Linearly convert the discrete coordinates of the matrix elements into geographic location coordinates; Based on the aforementioned geographical coordinates, and using a weighted calculation based on population density and building asset value, the risk level of casualties among ground personnel is quantified to obtain the population density layer. Based on the geographical coordinates, the ability of surface shelters to protect people on the ground is quantified to obtain the shelter layer; Based on the geographical coordinates, the height distribution of obstacles is quantified to obtain the obstacle layer; Based on the aforementioned geographical coordinates, the distribution of no-fly zones is quantified to obtain no-fly layers; Based on the drone's cruising altitude, the population density layer, the shielding layer, the obstacle layer, and the no-fly zone are merged to generate an initial ground risk map that includes the test flight zone, the restricted flight zone, and the no-fly zone.

3. The method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios according to claim 1, characterized in that, Constructing a horizontal ground risk field includes: Determine the horizontal flight direction angle of the drone and quantify the collision risk when the drone flies horizontally along the edge of the no-fly zone and restricted flight zone; Using UAV positioning error, size and mass, obstacle positioning error, and wind field influence as core indicators, the coordinates of the target obstacle and the UAV position are obtained. The positioning error is determined to follow a normal distribution. The distance error and standard deviation are calculated from the target obstacle position coordinates and the UAV position coordinates. A simplified standard deviation is obtained based on the standard deviation of the obstacle positioning error and the standard deviation of the UAV positioning error. By combining the collision probability, wind field correction factor, simplified standard deviation, maximum radius of the UAV, and equivalent radius of the obstacle, the horizontal safety interval is obtained; The no-fly zone and restricted-fly zone are horizontally dilated, with the dilation scale based on the horizontal safety interval. Multiple risk levels are divided, and a monotonically decreasing scale function of the horizontal distance from the current point to the nearest no-fly or restricted-fly boundary is solved to obtain the horizontal ground risk field.

4. The method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios according to claim 1, characterized in that, Constructing a vertical ground risk field includes: Determine the vertical flight direction angle of the drone and quantify the collision risk when the drone flies vertically along the edge of the no-fly zone and restricted flight zone; A dual-objective optimization model constrained by multiple conditions is constructed based on the drone boundary crossing conflict rate, buffer airspace ratio, cruising speed, size and mass, distance and wind field influence. The vertical safety interval is obtained by solving the dual-objective optimization model. The dual-objective optimization model includes the optimization objective of minimum boundary collision rate and the optimization objective of minimum buffer space ratio. The multiple constraints include deceleration distance condition, collision judgment condition, positioning error probability condition, velocity and acceleration constraint condition, and safety interval value constraint condition. The no-fly zone and restricted-fly zone are data-dilated in terms of vertical distance. The dilation scale is based on the vertical safety interval and multiple risk levels are divided. The monotonically decreasing scale function of the vertical distance from the current point to the nearest no-fly or restricted-fly boundary is solved to obtain the vertical ground risk field.

5. The method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios according to claim 1, characterized in that, An anisotropic ground risk field is constructed based on the horizontal ground risk field and the vertical ground risk field, including: In the local coordinate system, the horizontal and vertical directions are determined to be mutually orthogonal principal axes, wherein the horizontal direction is parallel to the boundary, the vertical direction is perpendicular to the boundary, and the horizontal ground risk field and the vertical ground risk field are cosine complementary. Determine the adjustable shape parameters, and obtain the horizontal angle weight and vertical angle weight from the shape parameters, the angle between the UAV flight direction and the horizontal direction, the horizontal ground risk field and the vertical ground risk field; The directional coupling correction term is obtained from the angle between the UAV's flight direction and the horizontal direction, the coupling coefficient, the horizontal ground risk field, and the vertical ground risk field; The anisotropic ground risk field is constructed by combining the horizontal angle weight, the vertical angle weight, the directional coupling correction term, the horizontal ground risk field, and the vertical ground risk field. The anisotropic ground risk field is constrained by boundary condition constraints and diagonal fly-through constraints.

6. The method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios according to claim 1, characterized in that, The anisotropic ground risk field is converted into a ground risk vector field, including: The maximum upward direction of the anisotropic ground risk field in the plane is obtained by using gradients, wherein the gradients in the x and y directions are obtained by central difference on the grid. Determine a noise threshold, and constrain the maximum upward direction with the noise threshold to obtain a unit vector of the direction with the fastest wind direction growth; The ground risk vector field is obtained by fusing the anisotropic ground risk field and the unit vector of the direction of fastest wind growth.

7. The method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios according to claim 1, characterized in that, Real-time risk calculation for any flight direction is achieved using the ground risk vector field, including: The ground risk vector field is decomposed into eigenvalues ​​to obtain a first orthogonal eigenvector and a first eigenvalue, as well as a second orthogonal eigenvector and a second eigenvalue. The first orthogonal eigenvector corresponds to the direction of maximum risk, and the second orthogonal eigenvector corresponds to the direction of minimum risk. Calculate the cosine and sine values ​​of the angle between the UAV's flight direction and the horizontal direction to obtain the unit vectors of the UAV's flight direction and the horizontal direction; Based on the first orthogonal eigenvector, the first eigenvalue, the second orthogonal eigenvector, the second eigenvalue, and the unit vectors of the UAV's flight direction and horizontal direction, a fast interpolation calculation is performed to obtain the risk value of the UAV in any flight direction.

8. A system for assessing the anisotropic risk of unmanned aerial vehicle (UAV) operation in urban low-altitude scenarios, characterized in that, include: The acquisition module is used to acquire UAV parameters and flight environment characteristic data to construct an initial ground risk map; The first construction module is used to construct a horizontal ground risk field and a vertical ground risk field based on the initial ground risk map. The second construction module is used to construct an anisotropic ground risk field based on the horizontal ground risk field and the vertical ground risk field. The conversion module is used to convert the anisotropic ground risk field into a ground risk vector field. The calculation module is used to perform real-time risk calculations for any flight direction based on the ground risk vector field.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for assessing the anisotropic risk of UAV operation in urban low-altitude scenarios as described in any one of claims 1 to 7.

Images