Community low altitude-oriented aircraft path planning and traffic capacity evaluation system

Abstract

The invention relates to a community low altitude-oriented aircraft path planning and traffic capacity evaluation system. The system comprises an airspace facility modeling module, a facility relation judgment module, a hierarchical airspace segmentation module, a noise-speed-safe distance constraint calculation module, a dynamic obstacle marking module, a multi-target path planning module and a bottleneck identification and traffic capacity evaluation module. The method comprises the following steps: modeling airspace facilities, and judging a facility relationship based on spatial geometric attributes and function use requirements of the facilities; performing hierarchical grid division on the low altitude of the community, calculating a safety distance along the movement direction and a safety distance perpendicular to the flight direction, and performing obstacle marking on grids around the building based on the safety distances and the flight direction; and performing path search based on an improved A * algorithm, identifying a bottleneck region, extracting parameters, and calculating the maximum allowable speed of the bottleneck region and the total traffic capacity of the bottleneck region. The system can relieve environmental troubles such as safety compliance dispute and noise pollution caused by operation of the low-altitude aircraft.

Description

Technical Field

[0001] This invention belongs to the field of low-altitude airspace management technology, specifically relating to an aircraft path planning and traffic capacity assessment system for low-altitude airspace in communities. Background Technology

[0002] With increasing emphasis on the development of the low-altitude economy, research on low-altitude traffic management related to vertical takeoff and landing (VTOL) aircraft has been put on the agenda. Current low-altitude flight management lacks consideration of the multi-story buildings and noise pollution in urban low-altitude environments, as well as the high safety requirements of VTOL, resulting in inefficient use of airspace resources, poor dynamic adaptability, and a lack of effective management methods and system support for urban low-altitude traffic.

[0003] While current patents related to low-altitude traffic management are showing diversified technological development, some limitations still exist in areas such as refined airspace management, dynamic adaptation, and multi-target collaboration in community scenarios, as detailed below: (1) Existing patents are insufficient in terms of the refinement of airspace division and dynamic adjustment capabilities. They are mainly based on static planning and lack consideration for urban low-altitude environmental factors and traffic demand matching. For example, Chinese patent CN119811141A focuses on signal light control technology in low-altitude conflict areas, Chinese patent CN208014189U monitors the airspace environment through smart terminals, Chinese patent CN119314362A plans airways based on a cellular structure, with layering based only on flight direction, and Chinese patent CN119862243A considers mid-to-long-distance airspace layering modeling. However, none of the above patents have carried out refined modeling of low-altitude areas in communities, making it difficult to cope with the dense buildings and varying heights of airspace environments in communities. This results in low utilization efficiency of low-altitude areas and causes disputes regarding environmental noise and the safety of people below the flight path.

[0004] (2) Existing technologies mainly focus on individual path planning, lacking consideration for modeling and optimizing the spatiotemporal resource allocation of the overall low-altitude traffic demand in the community airspace. For example, Chinese patents CN120178910A, CN118605568A, and CN119964417A are used to optimize the trajectory of a single UAV to avoid collisions; Chinese patent CN114117322A studies the macroscopic capacity of the network through model design and heuristic algorithms, but lacks refined traffic capacity; and Chinese patent CN120213037A focuses on emergency rescue path planning. Therefore, it is necessary to design a community low-altitude traffic intelligent management and planning system. Summary of the Invention

[0005] In view of the shortcomings and deficiencies of the existing technology, the purpose of this invention is to provide an aircraft path planning and traffic capacity assessment system for low-altitude communities. When planning the path, it can ensure the flight efficiency of the aircraft, strictly control the flight noise and avoid the risk of building collisions, effectively alleviate the safety disputes and environmental problems of low-altitude flights in communities, and predict the risk of congestion in bottleneck areas in advance based on the obtained traffic capacity results, providing a quantitative basis for airspace scheduling.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A system for aircraft path planning and traffic capacity assessment for low-altitude communities includes an airspace facility modeling module, which uses a five-tuple to uniformly model all airspace facilities within the low-altitude community. The low-altitude community is an airspace defined by a delineated community plane range and a specified altitude range, which is dedicated to low-altitude aircraft that take off, land, or pass through the delineated community for transportation or cargo transport. The facility relationship determination module is used to determine facility relationships based on the spatial geometric attributes and functional usage requirements of the facilities. The hierarchical airspace segmentation module is used to perform hierarchical grid-based division of the low-altitude airspace of the community based on the building density or traffic complexity of the area, and to further recursively subdivide it using an octree in densely built areas. The noise-velocity-safe distance constraint calculation module is used to calculate the maximum permissible flight speed of aircraft in each grid in real time or in advance based on the noise-velocity function, the noise-distance function and the noise limits of different functional zones in the community, and further derive the safe distance along the flight direction and the safe distance perpendicular to the flight direction. The dynamic obstacle marking module is used to dynamically mark "no-passage zones" or "obstacle segments" on the grid or octagonal sub-nodes around the building based on the current flight direction, the safe distance along the flight direction, and the safe distance perpendicular to the flight direction. A multi-objective path planning module is used for path search within an improved A* algorithm framework; The bottleneck identification and capacity assessment module is used to identify bottleneck areas from multiple historical or real-time paths output by the multi-objective path planning module, extract bottleneck area parameters, calculate the maximum allowable speed of the bottleneck area, and calculate the total capacity of the bottleneck area based on the single-channel-parallel channel model.

[0007] As a preferred embodiment of the present invention, the upper limit of the low-altitude altitude of the community is set to any fixed or dynamic altitude value within the range of 0 to 150 meters, and the upper limit of the altitude and the horizontal boundary can be dynamically adjusted according to urban planning or temporary control orders.

[0008] As a preferred embodiment of the present invention, the airspace facility modeling module is used to uniformly model all airspace facilities within the low-altitude region of the community using a quintuple F = [S, V, D, T, f]; where S is the boundary of the airspace, V is the airspace enclosed by the boundary S, which is a three-dimensional space, D is the interface set, T is the set of marker points used to identify key locations or reference points in the airspace, and f is the facility function type, which is divided into container type, channel type, and interface type, including at least two of the following: container type facilities include dedicated airspace for logistics drone take-off and landing and dedicated airspace for VTOL manned shuttles, and channel type facilities include one-way or two-way dedicated flight channels; The facility relationship determination module mainly classifies facility relationships into three categories: inclusion relationship, adjacency relationship, and connectivity relationship.

[0009] As a preferred embodiment of the present invention, the hierarchical spatial segmentation module uses cubic grids with a grid side length of 0.5 to 3 meters in densely built-up areas, combined with at least octree subdivision, and uses cubic grids with a grid side length of 5 to 20 meters in open areas.

[0010] As a preferred embodiment of the present invention, the noise-velocity function The expression is: ;in, denoted as , where is the sound pressure at a distance of 10 meters from the center of the aircraft during flight; k is the aircraft type correlation coefficient, used to calibrate the aerodynamic noise differences between different aircraft types; D0 is the reference sound pressure at a distance of 10 meters from the center of the aircraft; v is the actual flight speed of the aircraft; and c is the local speed of sound. Noise-Distance Function The expression is: ; Noise at target point The expression is: ; in, The distance from the target point to the center of the aircraft. This is the theoretical correction term for converting sound power level to sound pressure level during spherical wave diffusion. Environmental degradation.

[0011] As a preferred embodiment of the present invention, a safe distance perpendicular to the flight direction. It is a circular surface: ;in, Noise limits for different areas ; safe distance along the flight direction At the same time, the impact of noise must be considered, and the maximum value of the two should be taken; among them, Minimum safe distance; This represents the real-time velocity of the aircraft within the grid. denoted as reaction time; a represents the maximum braking acceleration.

[0012] As a preferred embodiment of the present invention, when the aircraft's flight direction points towards a building, the dynamic obstacle marking module extends along the flight direction, using the building as a reference. The grid within the area is marked "No Entry"; when the aircraft's path needs to pass between two buildings, the grid extends perpendicularly to both sides of the path direction, using the line connecting the edges of the two buildings as a reference. Range; Connected regions are identified using clustering algorithms; if the passable width of a connected region is less than... If so, the connected region is marked as an "obstacle segment".

[0013] As a preferred embodiment of the present invention, the total cost function of the improved A* algorithm is f(n) = g(n) + h(n); Where g(n) is the actual cumulative cost from the starting point to the current node n; h(n) is the heuristic estimated cost from the current node n to the end point. g(n) =ω1·L(n) +ω2·T(n) +ω3·Np(n) +ω4·Sp(n) +ω5·Fp(n); Where L(n) is the basic cost, T(n) is the time cost, Np(n) is the noise cost, Sp(n) is the total safety cost, and Fp(n) is the cross-boundary cost; the parameters ω1, ω2, ω3, ω4, and ω5 correspond to the weights of the indicators L(n), T(n), Np(n), Sp(n), and Fp(n) respectively, which are used to weigh and synthesize the indicator costs that need to be considered when there are multiple conflicting objectives. The heuristic cost estimation h(n) = α·3D Manhattan distance + β·building proximity penalty + γ·functional adaptability; where α, β, and γ are all weighting coefficients; Building proximity penalty: If the distance from node n to the nearest building is less than D... ⊥safe If yes, add 50; otherwise, add 0. Functional adaptability is used to characterize the degree of consistency between the functional attributes of the airspace or channel where the candidate node is located and the current flight mission type: a node is deducted 20 when it is in a dedicated airspace, deducted 10 when it is near a dedicated channel, and prohibited or increased by 200 when the candidate node enters a dedicated airspace / dedicated channel that is not of this mission type.

[0014] As a preferred embodiment of the present invention, the bottleneck area parameters extracted by the bottleneck identification and traffic capacity assessment module include the closest horizontal distance D between the outer contours of the two buildings. clear Bottleneck length L bottleneck Bottleneck area available height H bottleneck Wherein, the bottleneck section length L bottleneckThe intersection of the main traffic path obtained from A* planning and the building outline is determined; Maximum permissible speed v in the bottleneck area allow = min { v noise v ⊥ v brake}, where: v noise It is the maximum speed obtained from the noise limit constraint; v ⊥ It is the maximum speed obtained from the lateral safety distance constraint; v brake It is the maximum speed obtained from the forward braking safety constraint.

[0015] As a preferred embodiment of the present invention, after determining the maximum permissible speed in the bottleneck area, the bottleneck identification and traffic capacity assessment module abstracts the bottleneck area into a "single-channel section" low-altitude channel. It first calculates the time for a single aircraft to pass through the bottleneck segment and the minimum time interval between adjacent aircraft to determine the single-channel traffic capacity; then, it calculates the total traffic capacity of the bottleneck area based on the number of parallel channels.

[0016] Advantages and beneficial effects of the present invention: (1) This invention addresses the low-altitude flight needs within community airspace by constructing an intelligent management system that includes three-dimensional spatial modeling, hierarchical path planning, and traffic capacity assessment. It is applicable to scenarios such as unmanned aerial vehicle logistics, vertical take-off and landing aircraft (VTOL), and emergency rescue.

[0017] (2) The present invention provides refined airspace modeling, which can efficiently adapt to the complex environment of the community and improve resource utilization: by defining airspace facilities through a quintuple (S,V,D,T,f), and combining hierarchical grid (high resolution of 2m×2m×2m in densely built areas and low resolution of 10m×10m×5m in open areas) and octree segmentation technology, the outline of buildings, no-fly zone boundaries and sub-airspace functional ranges are accurately captured, solving the problem of coarse low-altitude modeling of communities in the existing technology; at the same time, it realizes the differentiated allocation of airspace resources, reduces the invalid airspace occupation, and significantly improves the utilization efficiency of low-altitude resources in the community compared with static planning.

[0018] (3) The path planning of this invention is multi-objective collaborative, taking into account both flight efficiency and safety and environmental protection requirements: It integrates noise-speed-safe distance closed-loop constraints (related to the Class II area limit of the "Environmental Quality Standard for Noise") and improved A* algorithm, and incorporates noise exceeding the standard penalty, safe distance violation penalty, and functional airspace cross-boundary penalty into the cost function, which not only ensures the flight efficiency of the aircraft, but also strictly controls flight noise (such as dynamically adjusting the lateral safe distance to 20m at night) and avoids the risk of building collision, effectively alleviating the safety disputes and environmental problems of low-altitude flight in the community.

[0019] (4) The present invention provides more accurate capacity assessment, supporting scientific scheduling and congestion risk prediction: The system can automatically identify the bottleneck channel segment with the most concentrated traffic flow from multiple historical or real-time planned paths, extract geometric parameters such as the nearest horizontal distance between buildings on both sides of the segment, channel length, and available height, and calculate the maximum allowable flight speed in the bottleneck area by combining noise constraints, lateral safety distance constraints, and braking distance constraints. It also establishes a "single channel → parallel multi-channel" capacity model with reference to road traffic flow theory to accurately quantify the theoretical peak capacity of the bottleneck area. The quantification results can be directly used for traffic monitoring and congestion early warning. When the predicted demand exceeds the capacity threshold of 70% to 90%, the system automatically triggers control strategies such as time-segmented take-off and landing, route diversion, or ground waiting, thereby achieving the orderly operation of low-altitude traffic in the community.

[0020] (5) This invention has strong scenario adaptability, is compatible with multiple business formats, and has broad expansion value: By functionally dividing the low-altitude airspace of the community (e.g., setting up dedicated take-off and landing sub-airspaces for logistics drones, dedicated shuttle sub-airspaces for manned VTOLs, and public flight channel sub-airspaces), designing one-way or two-way connectivity rules, and configuring dedicated interfaces, the system naturally supports the safe and efficient collaborative operation of multiple business formats such as drone logistics, manned low-altitude shuttles, emergency rescue, and medical supply delivery within the same community's low-altitude airspace. The core technology framework and algorithm are highly decoupled from the specific community scale and building distribution, and can be directly transferred to residential communities of different cities, densities, and altitudes, providing a replicable technical solution for the standardized and large-scale implementation of the low-altitude economy in urban communities. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the model built by the airspace facility modeling module of the present invention; Figure 2 This is a schematic diagram for automatically dividing an octree into slices; where a) an octree occupies a 25m slice; b) an octree occupies a 15m slice; c) an octree occupies a 5m slice; Figure 3 Schematic diagram of the obstacle region of the present invention Figure 1 (45m layer grid slice); Figure 4 Schematic diagram of the obstacle region of the present invention Figure 2 (25m layer grid slice); Figure 5 Schematic diagram of the obstacle region of the present invention Figure 3 (15m layer grid slice); Figure 6 This is a schematic diagram of the low-altitude airspace facilities in the community according to the present invention; Figure 7 This is a schematic diagram of the extended nodes for the low-altitude flight path planning of the present invention for communities. Figure 8A schematic diagram of the 3D path planning results Figure 1 (Top view); Figure 9 A schematic diagram of the 3D path planning results Figure 2 (Front view); Figure 10 A schematic diagram of the 3D path planning results Figure 3 (Side view); Figure 11 This is a 3D stereoscopic diagram of the community's low-altitude airspace and route planning according to the present invention. The three-dimensional coordinate system adopts a right-handed coordinate system with the southwest corner of the community as the origin. The X-axis represents the horizontal distance in the east-west direction, the Y-axis represents the horizontal distance in the north-south direction, and the Z-axis represents the vertical height relative to the ground. The unit of all three axes is meters. Figure 12 This is a flowchart of the processing flow of the low-altitude aircraft path planning and traffic capacity assessment system for communities according to the present invention. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions and advantages of the present invention, the present application will be described in detail below with reference to the accompanying drawings, but this is not intended to limit the scope of protection of the present invention.

[0023] This embodiment introduces the concept of "Community Low Altitude" (CLA), which refers to a bounded airspace defined by a demarcated community plane range (one or more defined horizontal ranges) and a specified altitude range. It is only for unmanned or manned vertical takeoff and landing (VTOL) aircraft to fly within the community. VTOLs coming from outside or leaving the CLA must enter and exit through the designated exit and entrance locations at the boundary of the airspace. The Community Low Altitude is exclusively for low-altitude aircraft that take off, land, or pass through the designated community for transportation or cargo transport.

[0024] This embodiment uses a high-density urban residential community as the application scenario. This community is a typical mixed-use residential area of ​​multi-story and high-rise buildings, requiring the coordinated low-altitude management of drone logistics delivery and small vertical take-off and landing (VTOL) aircraft for manned shuttle services. Specific scenario parameters are as follows: CLA (Community Low Altitude Area) zone: The horizontal boundary is rectangular, with an east-west length L = 500m and a south-north width W = 300m (horizontal boundary: x0 = 0, y0 = 0, w = 500, h = 300); the vertical height range is 0-100m (vertical boundary: Z = 0~100m). The airspace volume of the community low altitude area = V = 500m × 300m × 100m = 1.5 × 10⁻⁶ m. 7m³; It includes 12 buildings (four 12-story residential buildings with a roof height of 54 meters, four 12-story residential buildings with a roof height of 36 meters, and 4 six-story supporting buildings with a roof height of 18 meters. The specific locations and numbers of each building are shown in Table 1. In this embodiment, the plane positions, floor heights, and numbers of the buildings are schematic data fictitiously set for ease of explanation and do not correspond to the specific building arrangements in the real community. However, the spatial characteristics of typical high-density urban communities are referred to in terms of scale, spacing, and density, which can be used to verify the effectiveness of the path planning and traffic capacity evaluation method.).; There is a community center square at the bottom coordinates (250, 150, 0), which serves as the community positioning point; there are 2 external airspace interfaces. The east side (500, 150, 15) is the logistics UAV entrance d1, and the southwest side (0, 150, 15) is the VTOL exit d2.

[0025] Table 1 Specific locations and numbers of each building; floor height Building Number Planar coordinates (X1, X2, Y1, Y2) 54m (18 floors) B1 (68, 93, 210, 230) 54m (18 floors) B2 (408, 433, 210, 230) 54m (18 floors) B3 (88, 113, 250, 270) 54m (18 floors) B4 (388, 413, 250, 270) 36m (12 floors) B5 (85, 115, 68, 93) 36m (12 floors) B6 (385, 415, 68, 93) 36m (12 floors) B7 (220, 260, 90, 130) 36m (12 floors) B8 (220, 260, 170, 210) 18m (6 floors) B9 (50, 80, 140, 160) 18m (6 floors) B10 (350,370, 200,220) 18m (6 floors) B11 (140, 160, 273, 288) 18m (6 floors) B12 (340, 360, 273, 288) Coordinate description: In Table 1, "plane coordinates (X1, X2, Y1, Y2)" are used to describe the axis-aligned outer contour rectangle range of the building on the community horizontal plane (z = 0). The present invention adopts a right-handed coordinate system with the southwest corner of the community as the origin: the X-axis is the horizontal distance in the east-west direction (positive eastward), the Y-axis is the horizontal distance in the north-south direction (positive northward), and the unit is meters. For any building Bi, the horizontal projection area it occupies is x ∈ [X1, X2], y ∈ [Y1, Y2] (where usually X1 < X2, Y1 < Y2). If the building is regarded as a volume vertically stretched from the ground, its three-dimensional occupied space can be expressed as: x ∈ [X1, X2], y ∈ [Y1, Y2], z ∈ [0, Hi]; where Hi is the roof height of the building (given by "floor height / number of floors"). In addition, the plane center point of the building can be written as ((X1 + X2) / 2, (Y1 + Y2) / 2), and the plane sizes are (X2 - X1) and (Y2 - Y1) respectively, which are convenient for subsequent calculation of geometric parameters such as building spacing, passage net width, and bottleneck section.

[0026] Type of aircraft: Logistics UAV: Static noise D 0物流 = 55 dB (at 10 meters), the maximum flight speed v of the aircraft 物流MAX = 40 m / s, the maximum braking acceleration a = 10 m / s 2 .

[0027] Small vertical takeoff and landing vehicle (VTOL): Static noise D 0VTOL = 55 dB (at 10 meters), the maximum flight speed v of the aircraft VTOLMAX = 40 m / s, the maximum braking acceleration a = 10 m / s 2 .

[0028] like Figures 1 to 12 As shown, based on the above scenario, this embodiment provides an aircraft path planning and traffic capacity assessment system for low-altitude communities. This system, based on the concept of low-altitude communities, introduces hierarchical (including octree) grid partitioning to model the airspace and facilities in low-altitude communities, improving the design of multi-objective A / B navigation. * The path algorithm completes the path planning and then evaluates the corresponding traffic capacity of the bottleneck area.

[0029] Specifically, the aircraft path planning and traffic capacity assessment system for low-altitude communities includes: an airspace facility modeling module, a facility relationship determination module, a hierarchical airspace segmentation module, a noise-speed-safe distance constraint calculation module, a dynamic obstacle marking module, a multi-objective path planning module, and a bottleneck identification and traffic capacity assessment module; such as Figure 12 As shown, the overall process of the system of the present invention includes the following steps: Step 1: Input basic data and scene collection for low-altitude areas of the community, including community boundary S, building outlines, interface set D, marker point set T, and aircraft type and performance parameters, etc. Step 2: The airspace facility modeling module uses a quintuple F=[S,V,D,T,f] to uniformly model various facilities within the low-altitude airspace of the community. Here, S is the airspace boundary, V is the airspace enclosed by boundary S (a three-dimensional space), D is the interface set, T is the set of marker points used to identify key locations or reference points in the airspace, and f is the facility function type, categorized as container, channel, or interface. The low-altitude airspace itself is defined as a special case of the quintuple and is defined as a global container facility. Step 3: The facility relationship determination module automatically determines topological relationships such as "inclusion relationship" and "connectivity relationship" based on the facility boundary and spatial location relationship, forming a hierarchical structure of "global airspace - sub-airspace - channel - interface"; Step 4: The hierarchical airspace segmentation module divides the low-altitude airspace of the community into layers based on the complexity of the area. Low-resolution grids are used for public flight channels V3, while high-resolution grids are used for dedicated airspace V1 for logistics take-off and landing, dedicated airspace V2 for VTOL connection, and the area around buildings. In densely built areas, octrees are used to further refine the subspaces. Step 5: The noise-speed-safe distance constraint calculation module calculates the maximum permissible flight speed and the safe distance along the flight direction for each grid or octree subspace, taking into account noise limits, maximum braking deceleration, and inter-building geometric relationships. and safe distance in the vertical flight direction And write it into the raster attribute vector; Step 6: The dynamic obstacle marking module determines the obstacle based on the current flight direction and... , Dynamically mark "no-entry zones" or "obstacle segments" on the grid or octree child nodes around the building, and explicitly map the safety distance constraints to the grid state. Step 7: The multi-objective path planning module runs the improved A* algorithm on the above navigation map, taking into account actual cost, heuristic cost, functional adaptability and safety cost, and solves the optimal multi-objective path between take-off and landing points or interfaces under the premise of satisfying noise and safety constraints. Step 8: The bottleneck identification and traffic capacity assessment module automatically extracts the bottleneck channel segments at the intersection of the main traffic path and the building outline, calculates the maximum allowable speed, single channel and parallel multi-channel traffic capacity of the bottleneck area, and uses the results for traffic monitoring and congestion risk warning.

[0030] In this embodiment, the airspace facility modeling module is the core of the entire system. It uses a quintuple F=[S,V,D,T,f] to uniformly model all airspace facilities within the low-altitude region of the community, representing all facilities in the airspace. The parameters are described in detail below: S (Boundary): Represents the boundary of the airspace. It can be a specific geometric boundary, such as the exterior wall of a building or the fence of a community, or it can be defined by a virtual boundary. For example, S can be a polygon defining the boundary of a community. Specifically, this embodiment uses a relative coordinate system, with the southwest corner of the community as the origin (0,0,0), denoted as O. The geometric boundary of region S includes a planar boundary, a vertical boundary, and a virtual constraint boundary. The planar boundary is a rectangle with vertex coordinates of (0,0,0), (500,0,0), (500,300,0), and (0,300,0); the vertical boundary is z=0~100; the virtual constraint boundary is a building outline buffer zone, i.e., a no-fly zone formed by offsetting the exterior walls of each building by 3 m, which can be adjusted according to relevant regulations and planning requirements. In this relative coordinate system, the spatial position of any point can be represented as (x, y, z), where x represents the horizontal distance of the point along the X-axis relative to the origin O, y represents the horizontal distance along the Y-axis relative to the origin O, and z represents the vertical height of the point relative to the ground (the z=0 plane). All three are in meters.

[0031] V (Airspace): Represents the airspace enclosed by the boundary S. V is a three-dimensional space that defines the volume or capacity of the airspace (low-altitude area of ​​the community). It can refer to a specific airspace area for use by drones, vertical take-off and landing aircraft, etc. For example, if the boundary is S and the airspace height is 100 meters, then the volume of V can be expressed as the planar area of ​​the community multiplied by the height of 100m. Specifically, in this embodiment, the global airspace V is further divided into 3 sub-airspaces: V1 represents the logistics drone take-off and landing sub-airspace (dedicated airspace for logistics take-off and landing), with the core take-off and landing point of the logistics drone (single point coordinates: (100,150,5)) as the functional center. The sub-airspace range is defined by the spatial diagonal vertex coordinates (90,140,0) and (110,160,10): horizontal direction X-axis 90~110m, Y-axis 140~160m (forming a 20m×20m area, around the core take-off and landing point to adapt to drone hovering adjustment, cargo loading and unloading and other take-off and landing operations), vertical direction Z-axis 0~10m (covering the height of the core take-off and landing point, while reserving vertical take-off and landing redundancy space to avoid conflict with ground facilities), volume 20×20×10=4000m³; V2 represents the VTOL docking sub-airspace (dedicated airspace for VTOL docking). The first airspace, centered on the VTOL core connection point (single-point coordinates: (360,210,20), connecting interface d4), is defined by the coordinates of the diagonal vertices (350,200,18) and (370,220,28). The second airspace is defined by the horizontal X-axis (350~370m), Y-axis (200~220m, forming a 20m×20m area suitable for VTOL passenger pick-up and drop-off operations), and vertical Z-axis (18~28m, fitting the roof height to avoid scratching roof facilities, while reserving vertical take-off and landing redundancy). Its volume is 20×20×10=4000m³. The third airspace, V3, represents the public flight corridor airspace, encompassing the entire community low-altitude (CLA) area (500m×300m horizontally, 0~100m vertically, total volume 1.5×10⁻⁶). 7 m 3 After deducting V1, V2, and the no-fly zone for buildings, approximately (approximately 1.05 × 10) 7 m 3 The remaining space after that has a volume of approximately 4.492 × 10⁻⁶. 6 m 3 .like Figure 6As shown in the diagram, the blue bar area is V1, representing the sub-airspace for logistics drone take-off and landing; the green bar area is V2, representing the VTOL passenger docking sub-airspace; and the remaining airspace not occupied by buildings is the public flight channel airspace V3. The blue V1 area is arranged around the core take-off and landing point of the logistics drone, mainly used for drone hovering adjustment, take-off and landing, and cargo loading and unloading. It is characterized by its low altitude, small horizontal range, and connection to the external interface d1. The green V2 area is set up on the roof of the supporting building, mainly used for VTOL passenger pick-up and drop-off and ground docking. It is characterized by its high proximity to the roof, connection to interface d4, and maintaining a safe distance from residential buildings.

[0032] D (Interface Set): Represents a set of interfaces. This includes interfaces located on boundary S or any sub-facility boundary (such as sub-airspace / channel boundary), used to connect facilities or facilities with external airspace. Interfaces can be physical interfaces and play a crucial role in airspace management, directly affecting airspace structure division and aircraft traffic flow design. They must be specifically designed based on the Community Low Altitude Area (CLA) airspace boundaries, sub-airspace distribution, aircraft functional requirements, and overall management logic, rather than being arbitrarily set. In this invention, to unify the geometric description and functional representation of interfaces, a single interface is represented by a binary tuple of "identifier point – interface carrier". Specifically, the interface set is denoted as D = [d1, d2, ..., dn], where each interface di consists of a set of identifier points. With interface carrier plane Determined jointly, denoted as Among them, the marker points The spatial coordinates of the interface's geometric center are used for rapid interface location during path planning and state updates; the interface carrier plane. This indicates the boundary opening region where the interface is located, used to determine the geometric relationship between the interface and the spatial boundary S and sub-spatial region V. The "Coordinates" column in the table provides the corresponding marker points for each interface. The coordinate values. For example, when the interface is a rectangular opening, the rectangular area it contains can be denoted as... And select the geometric center of the rectangle as the marker point. When the interface is a circular or elliptical opening, the circular / elliptical area it occupies is denoted as... Select its center / ellipse center as the marker point. The formal definition of the marker set and its role in path planning and capacity assessment will be further explained in the definition section of the marker set T later. Specifically, this embodiment sets four interfaces, and the interface set... The specific functions are as follows: Table 2 Interface Number Table; Interface number coordinate type Function Description d1 (500,150,15) physical interface Logistics drones entering and exiting CLA entrances and exits [d2] (0,150,15) physical interface VTOL exit from CLA [d3] (100,150,5) physical interface Connection point between V1 and public flight corridor airspace V3 [d4] (360,210,20) physical interface Connection point between V2 and public flight corridor airspace V3 T (Identifier Set): This is a set of identifiers used to identify key locations or reference points in the airspace. It provides positioning benchmarks for path planning, clarifies key nodes for airspace navigation, and supports bottleneck location and safety constraint verification in subsequent capacity assessments. The selection of identifiers must consider the scenario parameters of the Community Low Altitude Area (CLA) (such as building distribution and sub-airspace range), aircraft traffic routes, and technical requirements; it is not arbitrary. The set can include the geometric center of the airspace, core functional points of sub-airspaces, building avoidance benchmarks, and path bottleneck warning points. For example, T can cover the community space center, core points of takeoff and landing sub-airspaces, high-risk building avoidance points, and narrow passage warning points, precisely matching the safety and efficiency requirements of low-altitude airspace management. Specifically, this embodiment defines... Used for route planning reference; Table 3. Marker Point Design Table; Identification point number type coordinate Design basis Functional attributes T1 Community Geometric Center (250,150,15) Based on the low-altitude plane range of the community (500m east-west and 300m south-north), the plane coordinates are the midpoint of the range (500 / 2, 300 / 2), and the vertical coordinates are 15m, which are adapted to the core passage altitude of the public flight channel V3 and serve as the reference point for overall airspace positioning. As an auxiliary reference node for path planning, it is used to correct the overall path direction, prevent the path from deviating from the core access area of ​​the community, and ensure that the aircraft movement line matches the overall airspace layout. T2 Take-off and landing center (100,150,5) The object is the same as the "core take-off and landing point of logistics drone take-off and landing sub-airspace V1", with no difference in coordinates. Logistics drone take-off and landing sub-airspace V1 is the take-off and landing area (range (90,140,0)-(110,160,10)) with this point as the functional center. Here, T2 is the core functional point of this area, and the coordinates are completely consistent. The two have a subordinate relationship of "area-center". The interface d3 coordinate consistent with T2 is the key end reference of the logistics unmanned aerial vehicle path planning (connecting V1 and the public flight channel V3), which ensures the accurate docking of the path starting point / ending point and the core operation point of take-off and landing, and avoids the mispositioning operation of the aircraft in the take-off and landing area. T3 Key points for building avoidance (300,200,27) Based on the parameters of the 12-story building (roof height 36m) in the low-altitude area of ​​the community and the design of the aircraft passage height: the plane coordinate (300,200) is located in the dense area of ​​the 12-story building and on the main passage line of the aircraft; the vertical coordinate is 27m, which is both in the core passage area of ​​the aircraft and below the 12-story roof. As an "obstacle reference benchmark" in path planning, the algorithm must ensure that the horizontal distance between any point on the path and T3 is greater than or equal to the vertical safe distance, so as to avoid the aircraft from approaching high-risk building areas and directly constrain the safety boundary of the path. T4 Bottleneck warning point (263, 150,20) Based on the design of the low-altitude building spacing scenario in the community, the narrowest point of the passage is usually selected: the plane coordinates (263, 150) are located between two 12-story buildings, which is the natural narrow section of the public flight passage V3; the vertical coordinate of 20m matches the main passage height of the aircraft, which can accurately locate the bottleneck position of the path. 1. Path Planning: Using T4 as a reference, detect the width of surrounding passages. If the width is less than the safe distance, mark the area as an obstacle. 2. Capacity Calculation: Using T4 as the starting point, determine the bottleneck section range, providing core positioning basis for calculating the maximum permissible speed and bottleneck capacity. f(Facility Function): A quintuple for modeling airspace facilities One of the core elements is to clarify the core purpose and operational attributes of various facilities within the airspace, and to achieve refined management of facilities through "functional classification"—connecting the spatial attributes of facilities (such as the boundary of S and the sub-airspace range of V) and matching the usage needs of aircraft (such as the take-off and landing of logistics drones and VTOL passage), providing a basis for the "functional specialization" of subsequent path planning (different aircraft take corresponding channels) and the "attribute prediction" of conflict prevention.

[0033] Facility function types are divided into three categories based on "spatial form + usage scenario". The definitions and functions of each category are as follows: Container type (C): refers to facilities with fixed space volume that can accommodate aircraft for "stay-at-home operations". Its core function is to provide dedicated space for aircraft to take off, land, hover, load and unload cargo (logistics drones) or pick up and drop off passengers (VTOL), so as to avoid occupying public passageways and causing conflicts during operation. Channel type (P): refers to facilities with linear connectivity that allow aircraft to "move through". Its core function is to plan fixed flight routes for specific types of aircraft, clarify the connection path between the "starting point and the destination", and reduce airspace intersections between different types of aircraft.

[0034] In this invention, the "channel" refers to a three-dimensional strip-shaped sub-airspace extending continuously along a predetermined centerline within the low-altitude airspace of a community, possessing a certain longitudinal length, lateral width, and usable altitude range, used to facilitate the continuous passage of aircraft between different facilities. Geometrically, the channel is jointly determined by one or more path centerlines and the surrounding effective passage cross-sections that meet noise limits and safe distance constraints; functionally, it is used to connect container-type facilities with interface-type facilities, or between different container-type facilities, enabling the organized flow of aircraft. Depending on the service target, channels can be divided into public flight channels and dedicated flight channels, with dedicated flight channels only open to specific types of aircraft; depending on the direction of passage, channels can be further divided into one-way channels and two-way channels.

[0035] Specifically, let the channel set be denoted as For ease of formal description, each channel P... k It can be represented as ;in, ∈D represent the inlet and outlet interfaces of the channel, respectively; The spatial orientation of the passage within region S is described by a series of path center points arranged in an orderly manner according to the flight direction; Σk is the cross-sectional function corresponding to Γk, used to give the passable cross-section at each center point (such as the safe radius of a cylindrical passage, the floor height interval, etc.). Therefore, a passage can be understood as "a three-dimensional connected sub-spacespace jointly determined by a tubular passage and its two end interfaces". The interfaces belong to the passage and connect with other facilities (such as container-type airspace, external road networks, etc.) through the interfaces. For example, in a subsequent embodiment, the connected path corresponding to the dedicated logistics passage P1 is P1:d1→d3, that is, starting from the logistics entrance interface d1 on the west side of the community, passing through a series of pre-planned path center points and a tubular airspace formed by a safe cross-sectional function, and finally ending at the interface d3 on the container-type facility V1.

[0036] Interface class (d): refers to node-type facilities located at the boundaries of different facilities to realize "spatial connection". Its core function is to serve as a "transition point" for aircraft to switch between different facilities (such as entering a public flight channel from external airspace or entering a public flight channel from a sub-airspace), and to clearly define the "entry / exit" attributes of passage.

[0037] Specifically, in this embodiment, the facility function classification corresponds one-to-one with the airspace facilities (V sub-airspace, D interface) mentioned above. The classification criteria and function matching logic are as follows: 1. Container type (C): Corresponds to logistics drone take-off and landing sub-airspace V1 and VTOL docking sub-airspace V2 。

[0038] V1 (sub-airspace range: (90,140,0)-(110,160,10)): As a dedicated take-off and landing space for logistics drones, it can accommodate drone hovering and adjustment, cargo loading and unloading and other stationary operations. It meets the "fixed volume + stationary operation" attribute of "container type", so it is classified as C. V2 (subspace range: (350,200,18)-(370,220,28)): As the dedicated docking space for VTOLs, it can accommodate VTOLs hovering, passengers getting on and off, and other stationary operations. It has the core characteristics of a "container" and is therefore classified as C.

[0039] 2. Channel type (P): corresponding to logistics drone channel P1 and VTOL channel P2.

[0040] P1 (path: d1→d3): The entry interface of channel P1 is... The export interface is Γ1 and Σ1 together form a tubular passageway from d1 to d3. P1 connects the "logistics drone entrance / exit d1 (external airspace access point)" and the "V1 sub-airspace connection point d3". It is a dedicated linear path for logistics drones to enter V1 from the external logistics airspace, allowing one-way passage for drones. It meets the attributes of "linear connectivity + mobile passage" of channel-type facilities, so it is classified as P.

[0041] P2 (path: d4→d2): The entry interface of channel P2 is... The export interface is Γ2 and Σ2 together form a tubular passageway from d4 to d2. P2 connects "V2 sub-space connection point d4" and "VTOL exit d2 (external space access point)", and is a dedicated linear path for VTOLs to travel from V2 to the external mid-to-high altitude access space. It allows VTOLs to pass in one direction and has the core characteristics of a passageway facility, so it is classified as P.

[0042] 3. Interface class (d): corresponds to all interfaces To unify the geometric description and functional representation of interfaces, each interface di is uniquely determined by a "identifier point – interface carrier plane" tuple, i.e. .

[0043] d1 (External Logistics Inbound Interface): Identifier Point Located on the eastern boundary plane of CLA (the geometric boundary of S, x=500). Its interface carrier plane is denoted as... Let be a rectangular opening on the boundary plane x=500, for example, it can be defined as... ,by It serves as the geometric center. This interface connects the "external logistics airspace" and the "public flight corridor V3," acting as a transition node for logistics drones entering the CLA, thus conforming to the "spatial connection" attribute of interface-type facilities.

[0044] d2 (External high-altitude connection interface): Identification point Located on the eastern boundary plane of the CLA. Its interface carrier plane is denoted as... Let be a rectangular opening on the boundary plane x=0, for example, it can be defined as...

[0045] d3 (V1 – Channel Conversion Interface): Identifier Point It is located on the boundary of subdomain V1. Its interface carrier plane is denoted as... The contact area between V1 and the common flight path V3 can be defined, for example, as... by It is the geometric center and satisfies It borders V3. This interface is used to connect "Public Flight Channel V3" and "Container Facility V1". It is a transition node for logistics drones to enter V1 from the public flight channel and belongs to the interface facility category.

[0046] d4 (V2 – Channel Conversion Interface): Identifier Point It is located on the boundary of subdomain V2. Its interface carrier plane is denoted as... The contact area between V2 and the common flight path V3 can be defined, for example, as... ,by It is the geometric center and satisfies It is adjacent to V3. This interface is used to connect "Container Facility V2" and "Public Flight Channel V3". It is a transition node for VTOL to enter the public flight channel from V2 and belongs to the interface facility category.

[0047] In this embodiment, Figure 1 The constructed model is illustrated. Figure 1 The thick black line represents the planar boundary of the low-altitude airspace S of the community (vertices O, A, B, and C correspond to (0,0), (500,0), (500,300), and (0,300) respectively); rectangles of different gray levels represent buildings with different heights of 18 m, 36 m, and 54 m; the orange dashed rectangle represents the public passage V3; the triangles on the left and right sides represent the logistics entrance d1 and the VTOL exit d2 respectively; the blue and green interface symbols represent the V1 interface d3 and the V2 interface d4; and the yellow pentagram represents the bottleneck point T4.

[0048] In this embodiment, the facility relationship determination module is used to determine the core facility relationships based on the spatial geometric attributes of the facilities (such as coordinate range and boundary overlap) and functional usage requirements (such as aircraft access routes and interface access attributes). These relationships are mainly divided into three categories: containment relationship (B), adjacency relationship (A), and connectivity relationship (L). The definitions, implementation logic, and specific examples of each type of relationship are as follows: Among them, B (inclusion relationship): indicates that the spatial range of one facility completely covers (or surrounds) another facility, all coordinate points of the included facility fall within the coordinate range of the included facility, and the two share a partial boundary. It is used to clarify the hierarchical affiliation of "global airspace - local sub-airspace" and support subsequent differentiated airspace management (such as raster resolution adjustment).

[0049] In this embodiment, the global community low-altitude (CLA) airspace (denoted as V) and sub-airspaces V1 (logistics drone take-off and landing sub-airspace) and V2 (VTOL docking sub-airspace) form an inclusion relationship, i.e., V V1, V V2. Sub-airspace V1 (dedicated airspace for take-off and landing of logistics drones) is an axis-aligned cuboid airspace. Its spatial range is strictly defined as follows: X coordinate: 90m ≤ x ≤ 110m, Y coordinate: 140m ≤ y ≤ 160m, Z coordinate: 0m ≤ z ≤ 10m. It is equivalent to a cuboid enclosed by the following 8 vertices: (90,140,0), (110,140,0), (110,160,0), (90,160,0), (90,140,10), (110,140,10), (110,160,10), (90,160,10). Subspace V2 (VTOL manned shuttle exclusive space) is an axis-aligned cuboid space, whose spatial range is strictly defined as follows: X coordinate: 350m ≤ x ≤ 370m, Y coordinate: 200m ≤ y ≤ 220m, Z coordinate: 18m ≤ z ≤ 28m. It is equivalent to a cuboid enclosed by the following 8 vertices: (350,200,18), (370,200,18), (370,220,18), (350,220,18), (350,200,28), (370,200,28), (370,220,28), (350,220,28). The global community's low-altitude CLA (Clear Space Domain) is an axis-aligned cuboid with spatial ranges of 0m ≤ x ≤ 500m, 0m ≤ y ≤ 300m, and 0m ≤ z ≤ 100m. It is easy to prove that all coordinate points of V1 and V2 strictly fall within the coordinate range of the global CLA, satisfying the inclusion relation. .

[0050] A (Adjacency Relationship): The adjacency relationship describes the geometric and topological relationship between two facilities that are "closely adjacent but not mutually inclusive" on the spatial boundary. It describes whether the facilities can be geometrically connected through interfaces, focusing on the connection methods between different airspaces or passages and potential intersection risk areas. Let the three-dimensional geometric space point sets of the two facilities Fi and Fj be respectively... (That is, the set of all points occupied by the facility in three-dimensional space). The topological operator Int(.) is introduced to represent the interior of the set (the set of points excluding the boundary). This represents the boundary (outer contour point set) of the set. Let the set of interfaces associated with facilities Fi and Fj be denoted as . Based on the definition of D (interface set) above, this representation will be used in the following calculations of adjacency relationships and effective passage area. On this basis, facilities Fi and Fj are defined to have an adjacency relationship, denoted as A(Fi,Fj) = 1, if and only if the following conditions are met simultaneously: ,

[0051] st ; in, The interface height tolerance is used to ensure that interface docking occurs at the same (or adjacent) height level. The above condition means that: within the boundary adjacent area of ​​the two facilities, there exists a pair of interfaces belonging to Fi and Fj respectively, whose interface carrier areas overlap non-emptily, thus forming a dockable passage section; otherwise, A(Fi,Fj)=0.

[0052] Preferably, in addition to meeting the above interface connection conditions, further constraints can be imposed. , This means that the internal areas of the two facilities do not overlap, and their boundaries are in contact or adjacent contact, excluding situations such as "inclusion" or "interpenetration" that do not conform to the facility zoning logic. It should be noted that the core determination of the adjacency relationship is based on the effective docking of the interface carriers, and the above geometric topological constraints are optional enhancement conditions.

[0053] When the above conditions are met, Fi and Fj are said to have achieved spatial adjacency and connectivity through the interface, and their effective connection area is defined as the area of ​​the overlapping region of the interface carriers: Seff(Fi,Fj)=Area( ∩ When multiple pairs of interoperable interfaces exist between two facilities, further definitions can be made. The effective cross-sectional area of ​​the two facilities as a whole can be further defined as follows: , of which M ijLet Fi be the set of all dockable interface pairs between facilities Fi and Fj, used to characterize the effective connectivity cross-sectional area of ​​the two facilities as a whole. In one implementation, to reduce computational complexity, the above formula is expressed as the sum of the overlapping areas of each interface pair (i.e., no union deduplication is performed on the overlapping areas of different interface pairs). In the scenario of channel-type facilities, if channel P... k The three-dimensional tubular space is regarded as Ω Pk The adjacency relationship between public airspace and exclusive sub-airspace can be determined through their interface docking.

[0054] In this embodiment, channel P1 (dedicated channel for logistics drones, d1→d3) and channel P2 (dedicated channel for VTOL, d4→d2) form a shared passage segment within the public flight channel V3. The key positioning points of the public channel segment are given by the set of marker points T; in this example, the key connection points of the public segment include T1=(250,150,15) and the bottleneck warning point T4, etc., used to describe the connection of P2 and the channel guidance of P1 (corresponding centerline examples: d4→T1→d2, d1→T4→d3). Its adjacency logic is as follows: V1 and the public flight channel V3 communicate through an interface... Complete the adjacency, and Area( ∩ Since Seff(V1,V3) > 0, A(V1,V3) = 1; V2 and the common flight channel V3 communicate via an interface. The adjacency is completed, so A(V2,V3)=1. It should be noted that in the engineering implementation, the common segment can be equivalently represented as several key marker points and their neighborhood range within the common channel, and is not limited to the form of a single fixed endpoint pair.

[0055] L (Connectivity): This characterizes whether two airspace facilities (or airspaces) are reachable and connected via channels and interfaces. If a sequence of paths is available for aircraft to pass through, the two are considered connected. Connectivity directions are categorized as "bidirectional" (allowing round-trip travel between the two facilities) and "unidirectional" (allowing only one-way travel from one facility to the other). The direction is constrained by interface functional attributes and business usage scenarios to standardize aircraft entry and exit rules and prevent chaotic traffic flow.

[0056] In formal representation, let the global interface set be D = {d1, d2, ..., dn}. For any facility Fi, its associated interface subset is D(Fi). D. If facilities Fi and Fj communicate via interface d k ∈D(F i )and ∈D(F j If a valid passageway (or physical connection) is formed by a given condition, then its connectivity is defined as follows: ;in, Represents the global interface d k At facility F i Interface instances on the boundary Represents the global interface At facility F j Interface instances on the boundary; p is the connection direction identifier, when p=single it indicates unidirectional connection, when p=double it indicates bidirectional connection.

[0057] For passage-type facilities (such as one-way dedicated passages), the engineering implementation of one-way connectivity can be constrained by the rule that "the projection progress along the predetermined direction of travel of the passage remains monotonically constant." This prohibits reverse travel during node-by-node / grid-by-grid expansion, ensuring that the directional constraint of p=single is strictly satisfied. Furthermore, in the engineering implementation of grid-by-grid expansion, cross-facility passage usually requires the following conditions to be met simultaneously: the adjacency relationship between the two facilities is established, A(Fi,Fj)=1, and the current point and the target point are located within the interface neighborhood of their respective facilities (in this embodiment, the interface neighborhood threshold is th=8m), to ensure that cross-area passage occurs near the defined interface docking position.

[0058] V1 and External Airspace: The bidirectional sub-airspace V1 (the logistics drone take-off and landing sub-airspace) is connected to interface d1 through a dedicated channel P1. The function of d1 is defined as "the physical interface for logistics drones to enter the CLA" (i.e., the connection entrance and exit between the external logistics airspace and the CLA). From the perspective of usage scenarios, logistics drones need to "enter V1 from the external airspace via d1→P1 (unloading)" and also need to "fly from V1 via P1→d1 to the external airspace (to perform the next delivery)", which requires bidirectional passage. Therefore, V1 and the external airspace are bidirectionally connected.

[0059] V2 and external airspace: One-way connection (only out, no in). Sub-airspace V2 (VTOL connection sub-airspace) is connected to interface d2 through dedicated channel P2. The function of d2 is defined as "the physical interface for VTOL to leave CLA" (i.e., the connection exit between CLA and urban high-altitude connection point). From the perspective of usage scenario, VTOL only needs to "fly to the external high-altitude connection point via P2→d2 after completing the passenger transfer in V2", and there is no need to "enter V2 from the external airspace" (this embodiment does not design the CLA entrance and exit of VTOL). Therefore, V2 and external airspace are one-way connection (only out, no in).

[0060] In this embodiment, the hierarchical airspace segmentation module is used to perform hierarchical gridding of the low-altitude airspace of the community based on the building density or traffic complexity of the area, and to further recursively subdivide the densely built areas using an octree. Specifically, in this embodiment, rasterization and octree segmentation are two key spatial segmentation techniques in the airspace modeling process. They are used for global planning and local path optimization, respectively, to ensure the accuracy of path planning in low-altitude flight environments. Rasterization divides the airspace into multiple units with fixed spatial resolution. Each raster unit represents a region in the airspace, and the size of the raster can be dynamically adjusted according to the complexity of the region. Through rasterization, the resolution of the airspace can be adjusted at different levels.

[0061] In this embodiment, the resolution can be dynamically adjusted according to the complexity of the area; specifically, it is divided into two layers. The first layer of grid is used to cover a larger area, so the resolution is lower and it is suitable for the public flight channel V3 area, i.e., the open area. Number of grid cells: 50 × 30 × 20 = 30,000; The second grid layer is used for areas requiring higher precision, therefore it has a higher resolution and is suitable for V1, V2, and areas surrounding buildings, i.e., densely built-up areas. Number of grid cells: 10×10×5×2=1000.

[0062] In this embodiment, the number of grid cells in each layer can be calculated using the following formula: ; Where N is the number of grid cells in the layer, SZ is the horizontal length of the spatial domain, W is the width of the spatial domain, H is the height of the spatial domain, and Δx, Δy, and Δz are the sizes of the grid cells.

[0063] To facilitate program implementation and subsequent calculations, this embodiment defines a uniform indexing method for each grid cell. Specifically, in the first layer ( Indicates a coarse resolution layer. In a high-resolution layer, let the number of raster cells in the X, Y, and Z directions be respectively... Then any grid cell can be represented by a quadruple. Unique identifier; among which, The grid number in the X direction; The grid number in the Y direction; This represents the grid number in the height direction. The center coordinates of the corresponding grid cell can be determined by the following formula:

[0064] in These are the origin coordinates of the spatial domain at this level. The formal representation and indexing rules described above facilitate unified access and updates of the raster in modules such as path planning, noise-velocity-safety distance constraint calculation, and dynamic obstacle marking.

[0065] To facilitate path planning, dynamic obstacle marking, and traffic capacity assessment, this embodiment defines a unified state and attributes for each grid cell: a) Grid status ∈{Passable, Impassable, Obstacle Segment} The initial state is set to "passable". In the dynamic obstacle marking module, based on the current flight direction and the calculated safe distance along the flight direction and the safe distance perpendicular to the flight direction, the grid (or octree child node) that is close to the building or has insufficient passage width is updated in real time to "no passage" or "obstacle segment".

[0066] b) Raster attribute vector

[0067] in, The maximum permissible flight speed of the aircraft within this grid under noise-speed-distance constraints. To maintain a safe distance along the flight direction, The safe distance in the vertical flight direction is obtained by the noise-speed-safe distance constraint calculation module, which combines noise limits and braking performance. Indicates whether the grid is located within a dedicated subspace or a dedicated channel; The path search cost weights associated with this grid are used to calculate the actual cost g(n) and heuristic cost h(n) in the improved A* algorithm.

[0068] Through the above state and attribute definitions, each grid cell in the navigation map has a clear "reachable / unreachable" accessibility attribute, and also carries physical constraint information related to noise, safe distance and functional adaptability, providing a unified data foundation for subsequent dynamic obstacle marking, multi-objective path planning and bottleneck accessibility assessment.

[0069] For public air corridor V3 (open area), the number of grid cells on the X-axis is 500 / 10 = 50; the number of grid cells on the Y-axis is 300 / 10 = 30; and the number of grid cells on the Z-axis is 100 / 5 = 20. For V1, V2, and the area surrounding buildings (densely built areas), taking a single sub-airspace such as V1 as an example: with a horizontal area of ​​20m × 20m and a vertical area of ​​10m, the number of grid cells on the X-axis is 20 / 2 = 10; the number of grid cells on the Y-axis is 20 / 2 = 10; and the number of grid cells on the Z-axis is 10 / 2 = 5.

[0070] In this embodiment, octree modeling is a further fine-grained subdivision of the two-layer hierarchical grid (dense building area), dividing it into smaller subspaces. Its core purpose is to further capture microscopic details such as building outlines and no-fly zone boundaries on the basis of the two-layer high-resolution grid, so as to meet the high-precision requirements of "safe distance verification" in path planning.

[0071] Specifically, starting from the initial space, each subspace is recursively subdivided. After each division, the size of the subspace is reduced to one-eighth of the previous space. That is, at the nth level, the space is divided into... Each subspace, until the preset minimum size is reached. In this embodiment, V1 is used as an example, with an initial space. Minimum subspace volume Calculate the number of segmentation layers n that satisfy Thus, n=3, and the initial space is finally divided into 512 refined subspaces. The actual settings will be used for engineering implementation.

[0072] In this embodiment, a rasterization method is used to segment larger areas within the airspace, while octree segmentation is used for finer areas (starting with a densely built-up area unit in the second layer of the hierarchical raster as the initial segmentation space; from the initial space, each subspace is divided into eight equal-sized subspaces, i.e., the subspace volume is reduced to 1 / 8 of the previous size, until the subspace volume reaches a preset minimum scale). This optimizes computational resources while ensuring path planning accuracy. The raster resolution and the number of octree layers are dynamically adjusted according to the actual needs of the airspace and the complexity of the area. For example, smaller raster cells or more octree layers are used in densely built-up areas, while larger raster cells and fewer octree layers are used in open areas.

[0073] The above design approach is mainly based on the following factors: Densely built areas: including the logistics drone take-off and landing sub-airspace V1, the VTOL docking sub-airspace V2, and the surrounding areas of 12-story residential buildings, 12-story residential buildings, and 6-story supporting buildings. These areas have a large number of buildings with significant height differences, requiring precise capture of building outlines, no-fly zone boundaries, and sub-airspace operation ranges (such as the 20m×20m take-off and landing area of ​​V1). Low resolution will lose details, so smaller grids are used to ensure accuracy; Open areas: referring only to most of the public flight channel V3. These areas have no dense buildings and are mainly used for aircraft "straight-line passage". There is no need to capture complex details, so larger grids can reduce the total number of grids, reduce the amount of computation, and avoid wasting computing power.

[0074] In this embodiment, the noise-speed-safe distance constraint calculation module is used to calculate the maximum permissible flight speed of an aircraft in each grid in real time or in advance based on the noise-speed function, the noise-distance function, and the noise limits of different functional zones in the community, and further derive the safe distance along the flight direction and the safe distance perpendicular to the flight direction. Based on this, this embodiment uses the grid unit simultaneously as a search node for path planning, a carrier for dynamic obstacle marking, and a basic unit for bottleneck identification and traffic capacity statistics: the multi-objective path planning module searches paths with the grid center as the node under the improved A* algorithm framework; the dynamic obstacle marking module marks the grids or octree child nodes around buildings as "no-passage zones" or "obstacle segments" based on the maximum permissible speed and corresponding safe distance within each grid; the bottleneck identification and traffic capacity assessment module calculates the maximum traffic capacity of the bottleneck area based on the grid flow traversed by the path, combined with the maximum permissible speed and effective cross-section of the corresponding grid.

[0075] In this embodiment, noise is associated with speed and safe distance to form a static constraint closed loop. Noise-speed limits and speed-safe distance are derived to ensure that the aircraft meets noise requirements and ensures flight safety during flight.

[0076] Noise-velocity function: ; Among them, noise level (dB) is the sound pressure level at a distance of 10 meters from the center of the aircraft during flight; the reference noise D0 (dB) is the sound pressure level when the aircraft is stationary, and is the reference sound pressure level at a distance of 10 meters from the center of the aircraft; velocity noise function The relationship between aircraft speed and noise is given by the formula. In low-altitude communities, the dominant noise sources are boundary layer pressure fluctuations and turbulent noise. The noise increment formula is nonlinearly related to the square of the velocity.

[0077]

[0078] Right now

[0079] in, This is the model correlation coefficient, used to calibrate the differences in aerodynamic noise between different models. The coefficient is 2000 for UAVs and 2500 for VTOLs. This reflects the nonlinear increase in boundary layer pressure fluctuations with velocity. v Let be the actual flight speed of the aircraft (m / s), and c be the local speed of sound, taken as 340m / s.

[0080] Meanwhile, noise decreases with distance, and the sound pressure at the target point is given by the following formula: ; ; in, Noise at the target point; For noise-distance function, The distance from the target point to the center of the aircraft. This is the theoretical correction term for the conversion of sound power level to sound pressure level during spherical wave diffusion, and is usually taken as . According to this embodiment, the reference sound pressure is set at 10 meters. For ease of calculation, it is simplified to... 10dB; For environmental attenuation (such as air absorption, obstacle shading, dB), set to 0.

[0081] The distance between the aircraft and the target point (sensitive area) is r. According to the "Environmental Noise Quality Standard" (GB3096-2008), the noise limit for different areas is... Noise limits for Category II areas (mixed residential, commercial, and industrial areas) are set for daytime (6:00-22:00). ≤55dB, nighttime (22:00-6:00) ≤45dB; then the maximum flight speed The following formula can be used to deduce the result: ≤ ,Right now

[0082] The permissible flight speed (permitted speed) of an aircraft is the minimum of the aircraft's speed limit and maximum flight speed, and the aircraft travels at that speed.

[0083] After obtaining the speed limit, according to the dynamic safety distance formula: Find the safe distance along the direction of motion. ; in, Minimum safe distance (meters, m); The real-time velocity of the aircraft in the grid (m / s); is the reaction time, set to 0.5 seconds (used to simulate the reaction delay of the pilot or system); a is the maximum braking acceleration.

[0084] At the same time, when the aircraft is stationary, it also needs to maintain a certain distance from buildings, that is: ; ; ; Therefore, the safe distance along the flight direction is: .

[0085] The vertical direction of flight is related to speed. When an aircraft flies at speed v, the safe distance perpendicular to the flight direction is a circular surface: .

[0086] Based on the above formula, the safe distances for the two types of aircraft along the flight direction and along the vertical flight direction, as well as the permissible flight speed in each grid, can be calculated respectively.

[0087] In this embodiment, the scene, aircraft, and constraint parameters required for calculation are shown in Table 4.

[0088] Table 4 Core Parameter Updates (UAVs); parameter Values / Explanations <![CDATA[Reference noise D0]]> 55 dB (sound pressure level at 10 meters when stationary) Model calibration coefficient k 2000 Speed ​​of sound c 340m / s The closest distance r to the sensitive area 20m Other parameters <![CDATA[D S =10dB, ΔD=0dB, tr=0.5s, a=10m / s², daytime D MAX =55dB, nighttime D MAX =45dB]]> The process of inversely calculating the permissible flight speed v (in the scenario of flying towards a building) based on noise constraints is as follows: When flying towards a building, noise limits for sensitive areas surrounding the building must be strictly met, ensuring that D''≤D MAX Deducing the maximum flight speed: Core formula: D0 + k·(v / c)² - [20lg(r) + D S +ΔD]≤D MAX。

[0089] Daytime scene (flying towards the building, D) MAX =55dB) Substitute parameters: 55+2000·(v / 340)²-[20 / lg20+10+0]≤55 Simplify: 2000·(v² / 115600)≤20×1.301+10=36.02(lg20≈1.301) v²≤(36.02×115600) / 2000≈2082→v≤45.6m / s Limited by the maximum physical speed of the drone (usually v) MAX =40m / s), take the allowable speed v=40m / s.

[0090] Night scene (flying towards the building, D) MAX =45dB) Substitute parameters: 55 + 2000·(v / 340)² - 36.02 ≤ 45 Simplify: 2000·(v² / 115600)≤45+36.02-55=26.02 v²≤(26.02×115600) / 2000≈1504→v≤38.8m / s At this point, the speed is allowed to be dominated by noise constraints, and the allowable speed is taken as v = 38.8 m / s (below the physical upper limit).

[0091] Safe distance D along the flight direction safe The calculation process (when flying towards a building) is as follows: Along the flight direction, both the "stationary noise safety distance" and the "dynamic braking distance" must be met simultaneously, and the maximum value should be taken.

[0092] Daytime scene: Static noise constraint safety distance D safe =0.316m, dynamic braking constraint (v=40m / s) safety distance D safe =100, take the maximum value: D safe =100m (Dynamic braking is dominant, ensuring braking redundancy when flying towards the building).

[0093] Nighttime scene: Static noise constraint safety distance D safe =1m, dynamic braking constraint (v=38.8m / s) safety distance D safe =94.7m, take the maximum value: D safe =95m (rounded to the nearest whole number to match the accuracy of path planning).

[0094] Safe distance D in the vertical flight direction ⊥safe The calculation process for (the two core values ​​when flying towards the building) is as follows: Vertical safety distance is a key obstacle avoidance indicator when flying towards buildings and must be directly linked to noise levels (ensuring noise compliance at the building site).

[0095] Flying towards the building during the day (v=40m / s), calculate Ma²=(40 / 340)²≈0.0138, k·Ma²=2000×0.0138≈27.6; Substitute into the formula: D ⊥safe ≈7.59m; Rounded (adapted to path planning grid accuracy): D ⊥safe =8m (minimum lateral distance when flying towards a building during the day).

[0096] Flying towards the building at night (v=38.8m / s), calculate Ma²=(38.8 / 340)²≈0.0130, k·Ma²=2000×0.0130≈26; Substitute into the formula: D ⊥safe ≈20m; Rounded down: D ⊥safe =20m (Minimum lateral distance when flying toward a building at night; the distance increases significantly due to lower noise limits).

[0097] In this embodiment, the dynamic obstacle marking module is used to dynamically mark the grid or octagonal sub-nodes around the building as "no-passage zone" or "obstacle segment" based on the current flight direction, the safe distance along the flight direction, and the safe distance perpendicular to the flight direction. Figures 3-5 This demonstrates the differences in the projected constraints of buildings on flyable space when 2D grid slicing is performed at different flight altitudes: the gray rectangles represent the building's floor plan, and the blue outlines represent the no-passage grid zones within that altitude level, formed by "building penetration of the level" and safety margins (obstacle expansion / grid quantization). As the slicing height decreases from 45m to 25m and then to 15m, more low- and mid-rise buildings begin to enter that altitude level and trigger the generation of no-passage zones. This results in a gradual increase in the number and coverage of blocked grids, a contraction of freely connected areas, and more pronounced narrowing and potential bottlenecks in passageways. This clearly reflects that the spatial constraints are stronger at lower altitude levels, and that passability is more easily affected by the distribution of building clusters.

[0098] This embodiment completes environmental modeling based on safe distance and flight direction. When constructing the navigation map, it fully considers the accessibility attributes and flight constraints of each grid level, and sets corresponding edge weights and reachability relationships in the navigation map to ensure the rationality and feasibility of path planning. At the same time, it integrates multi-objective planning pre-calculation to provide more efficient support for subsequent path planning.

[0099] Specifically, firstly, obstacle areas are modeled based on safety distance design constraints, combined with safety distances along the flight direction. (recorded as) ) and safety distance in the vertical path direction (recorded as) Obstacle marking is applied to the grid around the building: Before each step of path planning, the aircraft's flight direction is determined. When the aircraft's flight direction points to a building, that building is used as a reference point to extend along the flight direction. The grid within the area is marked as a "No Entry Zone" (to prevent aircraft from entering the danger zone when heading towards buildings); the forward safety distance D1 is consistent with the safety distance along the flight direction within the current grid, and is adaptively set using the following formula:

[0100] That is, before each path planning step, D is calculated based on the noise-speed-braking constraints. safe As a forward shielding distance, the grid within a range D1 in front of the building is uniformly marked as a "no-entry zone". Based on the drone scene parameters in this embodiment, under daytime conditions, D... safeThe distance is approximately 100 m, and approximately 95 m under nighttime conditions. Therefore, the typical value of D1 falls within the range of [90 m, 110 m]. In densely built-up areas, a lower limit truncation can be applied to D1 to balance safety and the utilization rate of passable airspace. Specifically, densely built-up areas use 2m×2m×2m high-resolution grids + octree subspace markers, while open areas use 10m×10m×5m low-resolution grid markers. Within this range, the grid is set to "No Entry".

[0101] When the aircraft's path needs to pass between two buildings, the line connecting the edges of the two buildings is used as a reference, extending perpendicularly to both sides of the path direction. Range. The lateral safety extension distance D2 is consistent with the vertical flight direction safety distance under the current operating conditions, and is set according to the following formula:

[0102] Connected regions are identified using clustering algorithms. If the passable width of a region (converted according to the corresponding grid step size) is less than 2× If so, the connected area is marked as an "obstacle segment" to ensure that the passage gap between the two buildings meets the safety requirements.

[0103] In this embodiment, during environmental modeling, the maximum permissible flight speed of the aircraft in all grids is first obtained, and the solution is then based on the maximum permissible flight speed. and And based on the calculation and Modeling is performed; using the above rules, the calculated safety distance is transformed into obstacle constraints in the grid map, thereby completing the environmental modeling before path planning, which facilitates subsequent path planning.

[0104] In this embodiment, the multi-objective path planning module is used to perform path search within the improved A* algorithm framework; This embodiment deeply integrates the multi-objective A* algorithm with the noise-safety modeling described above, forming a complete closed loop of "modeling constraints → algorithm implementation → path output". It highlights the rigid constraint role of safety distance and noise limits in A* search, and provides an example demonstration using a scenario of an aircraft (UAV) flying towards a building.

[0105] Multi-objective A* algorithm design: The core framework of the algorithm: The total cost function of the A* algorithm is f(n) = g(n) + h(n); Where g(n): the actual cumulative cost from the starting point to the current node n; h(n): the heuristic estimated cost from the current node n to the end point.

[0106] Definition of actual cost g(n): In high-risk scenarios where drones fly towards buildings, multi-dimensional cost constraints are needed to ensure flight safety, noise compliance, and functional compatibility. Therefore, the actual cumulative cost from the starting point to the current node n in the A* algorithm, g(n), must include the following constraint terms: g(n) = ω1·L(n) + ω2·T(n) + ω3·Np(n) + ω4·Sp(n) + ω5·Fp(n).

[0107] Table 5. Information on each parameter in the actual cost g(n); Cost items definition Weight ω (when flying towards a building) Basic cost L(n) The 3D path length (m) from the starting point to node n is accumulated by the center distance of the raster / octree nodes. ω1=2 (basic efficiency) Time cost T(n) Flight time (s) from the starting point to node n ω2=3 (time efficiency) Noise cost Np(n) <![CDATA[Noise over - standard penalty (dB): If the noise D'' at node n > D MAX , then Np = 10×(D'' - D MAX ); otherwise it is 0;]]> ω3=5 (Noise rigid constraint) Total security cost Sp(n) The total safety cost is the sum of the "lateral compression penalty" and the "forward collision risk penalty". ω4=10 (Safety rigid constraint) Cross-boundary cost Fp(n) Penalties are applied based on the match between node location and aircraft type: 1. Compliant Area (Penalty=0): The node is located within the dedicated facility for this type of aircraft (e.g., a logistics drone located in V1, P1, or public passageway V3). 2. Restricted / Cross-boundary Area (Penalty=1000): The node enters the dedicated airspace of other types of aircraft (e.g., a logistics drone mistakenly enters the passenger shuttle area V2 or VTOL passageway P2), which is a rigid restriction, and extremely high penalties are imposed to force avoidance. ω5=20 In this embodiment, time cost Total security cost ; Vertical (lateral) squeeze penalty The expression is:

[0108] in, The horizontal projected distance from the node to the nearest obstacle; Collision risk penalty in the forward direction The expression is:

[0109] in, It is the straight-line distance from the node along the direction of flight speed to the surface of the obstacle.

[0110] Weighting logic: When flying toward a building, safety (ω4) and noise (ω3) have the highest priority to avoid risks caused by approaching the building.

[0111] The definition of heuristic cost h(n): To guide the path away from buildings and quickly reach the destination, h(n) is designed as: h(n) = α·3D Manhattan distance + β·building proximity penalty + γ·functional adaptability; 3D Manhattan distance: |x n -x end |+|y n -y end |+|z n -z end |;where x n y n z n The coordinates of node n; x end y end z end The coordinates representing the endpoint.

[0112] Building proximity penalty: If the distance from node n to the nearest building is less than D... ⊥safeIf it is a danger warning zone, add 50; otherwise, add 0. Functional adaptability: Nodes within dedicated airspace are deducted 20 (encouraged), nodes near dedicated channels (such as P1) are deducted 10, and when a candidate node enters a dedicated airspace / dedicated channel that is not of this task type, it is prohibited from entering or deducted 200 (avoidance).

[0113] Coefficients: α=1.2 (distance weight), β=2 (building avoidance requirements), γ=1.

[0114] In this embodiment, functional adaptability is used to characterize the degree of consistency between the functional attributes of the airspace or channel where the candidate node is located and the current flight mission type. The system pre-divides each sub-airspace and channel in the low-altitude community into dedicated airspace for logistics drones, dedicated airspace for VTOL, and public shared airspace according to the service objects, and labels the corresponding functional types for different flight missions. When candidate node n is located in a dedicated airspace that perfectly matches the mission type, it is determined that its functional adaptability is high, and the functional adaptability item is set to -20 to reduce the heuristic cost of the node and encourage the path to unfold in the dedicated airspace first; when node n is located in the public airspace but is adjacent to the dedicated channel of the current mission at the grid scale, the functional adaptability item is set to -10 to guide the path to merge into the dedicated channel as soon as possible; when the airspace function corresponding to node n is inconsistent with the current mission type and there is a situation of crossing the functional boundary, it is determined as "crossing the boundary", and the functional adaptability item is set to +200, which significantly increases the heuristic cost of the node, thereby actively avoiding cross-boundary passage during the search process. The aforementioned weights were selected through multiple sets of simulation calibrations to guide A* search results to prioritize the use of dedicated airspace and channels for functional matching, while meeting safety and noise constraints, thereby reducing airspace overlap and potential conflicts between different business types.

[0115] Node expansion and constraint checking: In A* search, each node's 26 adjacent directions (including diagonals) must pass a four-level constraint check (priority from high to low). Nodes that fail the check are directly excluded, as detailed in Table 6. Table 6 Constraint Checks; Inspection level Inspection content Quantitative judgment criteria (taking logistics drones flying towards buildings during the day as an example) 1. Function-specific checks Is the node within the dedicated airspace / channel of the corresponding aircraft? The function-specific check is a hard constraint; if it fails, the candidate node is directly removed. 2. Flight direction check <![CDATA[Whether it is "towards the building". If it is towards the building, the safety distance D along the direction needs to be satisfied.]] safe > Building orientation determination: The angle between the flight vector (current → next node) and the vector "node → building center" is ≤90°; the distance along the direction is ≥100m. 3. Vertical safety inspection <![CDATA[The 3D vertical distance from the node to the building ≥ D ⊥safe > Lateral safety distance checks use a horizontal projection distance ≥8m 4. Speed ​​and noise check <![CDATA[The speed of the section where the node is located ≤ the allowable speed v, and the noise D'' ≤ D at this speed MAX > For a speed ≤ 40 m / s, D'' = 55 + 2000 × (40 / 340)² - [20lg(r) + 10] ≤ 55 dB The following example uses a logistics drone flying towards a building during the day (starting point d1(500,150,15), ending point d3(100,150,5), passing through bottleneck point T4(200,150,15)). The modified A* path search algorithm is executed, and the steps are as follows: Step 1 (Initialization): Based on the hierarchical grid and octree modeling results, read the access attributes and constraint parameters of each grid, set the starting point s and the ending point t, initialize the open list Open (containing only s) and the closed list Closed (empty), and set the weight coefficients of the cost functions g(n) and h(n).

[0116] Step 2 (Select the current node): When Open is not empty, select the node n* with the smallest total cost f(n) = g(n) + h(n) from Open as the current expansion node and move it into Closed; if n* has reached the endpoint t or the grid where the endpoint is located, then go to step 6.

[0117] Step 3 (Generate candidate nodes): Generate a set N(n*) of n* adjacent nodes in the 3D grid according to the 26 neighborhood rule, and remove static obstacles and nodes that have been marked as prohibited.

[0118] Step 4 (Constraint Filtering): For each candidate node n′∈N(n*), perform function-specific checks, flight direction checks, vertical safety distance checks, and velocity-noise constraint checks in sequence. Nodes that fail any of the constraints are discarded directly.

[0119] Step 5 (Cost Update): For node n′ that passes the constraint, update g(n′) according to the current motion increment, calculate h(n′) according to the heuristic function, obtain f(n′)=g(n′)+h(n′), and add or update it according to whether it already exists in Open / Closed, while recording the parent node as n*; then return to step 2.

[0120] Step 6 (Path Backtracking and Output): When the endpoint node is removed from Open, backtrack from the endpoint along the parent node pointer back to the starting point to obtain the optimal discrete path from s to t. Further trajectory smoothing and bottleneck analysis can be performed as input for subsequent capacity assessment.

[0121] In this embodiment, the bottleneck identification and traffic capacity assessment module is used to automatically identify bottleneck segments from multiple historical or real-time three-dimensional flight paths output by the multi-objective path planning module, extract key geometric and operational parameters of the bottleneck segments, calculate the maximum allowable speed of the bottleneck segments, and calculate the total traffic capacity of the bottleneck segments based on the "single channel - parallel channel" model.

[0122] 1) Path segment analysis and local coordinate construction For any candidate path P={p0,p1,…,p n}, where pi=(xi,yi,zi) is the coordinate of the i-th 3D path point, and n is the total number of path points minus 1. This module uses the path segment formed by adjacent path points as the analysis unit. For each segment (p i-1 ,p i ),calculate: 1. Midpoint m of the segment i m i =(p i-1 +pi ) / 2; Where, m i Let be the coordinates of the midpoint of the i-th segment; 2. Heading direction unit vector

[0123] Let Δx = x i+1 -x i , Δy=y i+1 -y i ,but ; in, Let i be the horizontal heading unit vector of the i-th segment; 3. Heading Normal Unit Vector

[0124]

[0125] in, To align with the course The vertical horizontal normal unit vector, They are respectively The x and y components.

[0126] 2) Velocity-Constraint Closed Loop: Local three-category velocity upper bounds and safe distance within m i The velocity-constraint closed-loop function is called at the point to obtain the noise constraint velocity, cross-sectional constraint velocity, and forward braking constraint velocity, and the minimum value of these three values ​​is taken as the local allowable velocity at that point.

[0127] 1. Noise-constrained speed v noise First, calculate the minimum horizontal distance r from the point to the outer boundary of the nearest building (considering building buffers). min , The noise propagation constraint is D''(v,r)=D(v)-[20lg(r)+D S +ΔD]; Where the velocity-noise relationship is D(v) = D0 + k × (v / c)²; The noise limit is D max (Day / night is determined by scene configuration), then D′′(v,r)≤D max The maximum speed under noise constraints is obtained as follows: ; in, This represents the upper limit of the aircraft's physical maximum speed; when the value within the square root is not positive, this upper limit is treated as 0.

[0128] 2. Cross-sectional constrained velocity v ⊥ (Interlocking between buildings) Along ± at mi If a ray is used to hit the building's outer contour from both sides, then the distance between the two hits is obtained. The net width of the cross-section between the buildings is: D clear,i =

[0129] in, For along + The distance from the direction to the nearest building facade; For along- The distance from the direction to the nearest building facade; D clear,i This is the actual horizontal clear width between the exterior facades of the two buildings; If no building is hit on either side (e.g., if a community boundary or open area is hit), then no building-to-building pair is formed (no building pair identifier is generated in this segment).

[0130] Based on the net width of this section, the usable half-width is defined (after deducting the half-width margin Δw): w half,i =max(0, 2D clear,i -Δw) Wherein, Δw is the cross-sectional half-margin / safety margin, used to deduct track deviations, control errors, and other margins.

[0131] and w half,i Substituting the lateral constraint scale into the inverse solution of the noise velocity yields the cross-sectional constraint velocity: ; Among them, v ⊥ This indicates the upper limit of velocity derived from the width of the lateral cross-section; 3. Forward braking constraint speed v brake along the course By performing a ray hit on the outer contour of the building in front, the distance L to the nearest obstacle in the forward direction can be obtained. f (where L) f (This refers to the distance from the current position along the heading to the nearest forward building facade). Braking safety constraints are based on: ; in, 'a' represents the reaction time; 'a' represents the maximum braking deceleration / upper limit of braking acceleration, in m / s². 2 ; The upper limit of braking is obtained as: If no building is hit in front, the constraint is not activated, and v is taken. brake = ; Combining the three types of velocity upper bounds and applying a minimum velocity constraint, the final velocity is v. allow : vallow = max( )) Among them, V min The system sets a minimum controllable speed to avoid numerical degradation; 3) Bottleneck center segment identification For all path segments, prioritize collecting the set of segments that satisfy the "building-to-building" constraint. I pair ={i | Buildings are hit on both sides of the cross section and a building pair (Ba,Bb) can be obtained}, where Ipair is the set of clamping segment indices; Ba,Bb are the building indices / numbers hit on the left and right sides of the clamping cross section; If Ipair ≠ Take,i * =arg ; where i * Index for the bottleneck center segment; This represents the net width between buildings calculated using the buffer building profile, used for robust identification of clamping relationships.

[0132] If Ipair= Then, speed is used as a safety net, i * = arg

[0133] in, Let i be the locally reachable velocity corresponding to segment i; The bottleneck center point is taken as the midpoint p of the bottleneck center segment. * = m i* 4) Bottleneck range expansion and bottleneck segment length When the bottleneck center segment belongs to the "building-to-building" type, the central segment building is used to pair with the other buildings. i* To constrain forward and backward expansion. Let the net width of the center segment (buffer profile) be... The extended threshold uses a scaling factor η>1 (which can be designed according to requirements; in this embodiment, η=1.15 is used): D thr =

[0134] Among them, D thr The upper limit threshold for the net width of the bottleneck range extension can be designed according to requirements.

[0135] If adjacent segments satisfy: pair j =pair i* ,and Then it will be merged into the bottleneck interval, and the final bottleneck interval will be... The bottleneck section length is:

[0136] Where i0,i n These are the start and end indexes of the expanded bottleneck interval, respectively. It is a three-dimensional Euclidean distance; This represents the total length of the bottleneck section. , These represent the coordinates of two adjacent bottleneck segments.

[0137] 5) Bottleneck parameter extraction Building pair identification: Record the building pair (Ba, Bb) at the center of the bottleneck section; actual horizontal clear width D between buildings. clear At p*, D is obtained by using the unbuffered, realistic building outline as the basis for ray tracing on both sides. clear = ;in, To be on the lower edge of the true contour ± The hit distance; the net width is used to explain the geometric nature of the bottleneck.

[0138] Effective net width D eff :D eff =min(max(0, D clear -2Δw), W design ); Among them, D eff W is the effective net width that can be used for parallel channel deployment. design To design the upper limit of corridor width; Height range H bottleneck H bottleneck =[z min ,z max ]; where z min , z max These represent the minimum and maximum heights of all points along the path, respectively.

[0139] 6) Bottleneck maximum allowable speed and source of dominant constraint: Take v at the bottleneck center point p* bottleneck =max( )) and with As the "dominant source of constraint," it is used to explain the causes of bottlenecks.

[0140] 7) "Single-channel - Parallel-channel" capability chain computation 1. Time t for a single aircraft to pass through the bottleneck period pass =L bottleneck / v bottleneck 2. Minimum time interval between adjacent aircraft (longitudinal anti-tailgating): t headway =D safe (v bottleneck ) / vbottleneck 3. Single-lane traffic capacity C single =3600 / (t pass +t headway (flights / hour) 4. Number of parallel channels When the effective net width is insufficient, i.e., n channel <1 indicates a 0-channel / unpassable condition.

[0141] 5. Total traffic capacity of the bottleneck area: C bottleneck =n channel ×C single .

[0142] In this embodiment, the longitudinal (rear-end collision prevention) safety distance is the maximum value of the static noise safety distance and the dynamic braking safety distance: D safe (v)

[0143]

[0144] Where, D safe (v) represents the minimum longitudinal safety clearance; This is the lower limit of the static safety distance derived from the noise threshold. Representing speed: Lateral safety distance (for parallel channel occupancy) is defined as:

[0145] Among them, D ⊥safe (v) represents the minimum lateral safety clearance.

[0146] Taking the example of a logistics drone passing through a bottleneck section during the day, let: D0 = 55 dB (reference noise constant), k = 2000 (aircraft noise figure), c = 340 m / s (velocity of sound), Dmax = 55 dB (daytime noise limit), D S =10 dB (spherical wave correction term), ΔD=0 dB (environmental attenuation), tr=0.5 (response time), a=10 m / s 2 (Maximum braking deceleration); D clear =34 m (actual net width between buildings), Δw=0.5 m (section half-width margin), then D eff =D clear -2Δw=33 m, L bottleneck =48 m (bottle section length): 1. Noise – Lateral Section Constraint Velocity: Determined by the actual net width D between buildings. clear=34 m and half-width margin of section Δw=0.5 m, the effective net width D is obtained. eff =D clear -2Δw=33 m; take the available half-width w half =D eff / 2=16.5 m; w half Substituting the lateral scale into the noise velocity inverse solution, the upper bound of the velocity v is obtained. ⊥ =v noise (w half ), we can get v ⊥ ≈44.6 m / s (approximately greater than 40 m / s). Therefore, lateral constraints are inactive, and the bottleneck velocity is dominated by other upper bounds (such as the physical speed limit of the aircraft type). Let v be... bottleneck =40 m / s.

[0147] 2. Braking constraint speed: The forward risk constraint in this segment is inactive, therefore v brake It is not a binding term, but the definition still exists; 3. Overall maximum permissible speed: v bottleneck =40m / s.

[0148] (3) Safety distance and time parameters Single aircraft transit time: t 通过 =1.2s; Interval between adjacent aircraft: t 间隔 =2.5s.

[0149] (4) Single lane and total capacity Single-lane capacity: C 单通道 ≈973 flights / hour; Number of parallel channels: n 通道 =2 (lateral safety distance accommodates two independent passages). Total capacity: C 瓶颈 =2×C 单通道 ≈1945 flights / hour.

[0150] The above-described specific examples are for illustrative purposes only and are not intended to limit the scope of the invention. Those skilled in the art can make various simple deductions, modifications, or substitutions based on the principles of this invention. Therefore, the scope of protection of this invention should be determined by the scope of the claims.

Claims

1. A community-oriented low-altitude aircraft path planning and capacity assessment system, characterized in that, The airspace facility modeling module is configured to uniformly model all airspace facilities in the community low airspace by using a five-tuple, wherein the community low airspace is an airspace defined by a designated community planar range and a specified height range, and is exclusively used for traffic or cargo transportation of low-altitude aircraft taking off or landing in the designated community; The facility relationship determination module is configured to determine facility relationships based on spatial geometric attributes and functional use requirements of the facilities; The hierarchical airspace segmentation module is configured to hierarchically rasterize and divide the community low airspace according to regional building density or traffic complexity, and further recursively subdivide the building-dense area by using an octree; The noise-velocity-safety distance constraint calculation module is configured to calculate the maximum allowable flight velocity of an aircraft in each grid in real time or in advance based on a noise-velocity function, a noise-distance function, and noise limits of different functional partitions of the community, and further derive a safety distance along a flight direction and a safety distance perpendicular to the flight direction; The dynamic obstacle marking module is configured to dynamically mark a "no-passing zone" or an "obstacle section" for a building peripheral grid or an octree child node according to a current flight direction, the safety distance along the flight direction, and the safety distance perpendicular to the flight direction; The multi-target path planning module is configured to perform path search under an improved A* algorithm framework; The bottleneck identification and traffic capacity evaluation module is configured to identify a bottleneck area from a plurality of historical or real-time paths output by the multi-target path planning module, extract bottleneck area parameters, calculate a maximum allowable velocity of the bottleneck area, and calculate a total traffic capacity of the bottleneck area based on a single-channel-parallel-channel model.

2. The community-oriented low-altitude aircraft path planning and capacity assessment system according to claim 1, wherein, An upper limit of the height of the community low airspace is set to be any fixed or dynamic height value in a range of 0-150 meters, and dynamic adjustment of the upper limit of the height and a horizontal boundary is supported according to urban planning or temporary control instructions.

3. The community-oriented low-altitude aircraft path planning and capacity assessment system according to claim 1 or 2, characterized in that, The airspace facility modeling module is configured to uniformly model all airspace facilities in the community low airspace by using a five-tuple F = [S, V, D, T, f]; wherein S is a boundary of the airspace, V is an airspace enclosed by the boundary S, which is a three-dimensional space, D is an interface set, T is a set of identification points for identifying key positions or reference points in the airspace, and f is a facility function type, which is divided into a container type, a channel type, and an interface type, and at least includes any two or more of the container type, the channel type, and the interface type; the container type facility includes a logistics unmanned aerial vehicle take-off and landing exclusive airspace and a VTOL manned connection exclusive airspace, and the channel type facility includes a one-way or two-way exclusive flight channel. The facility relationships in the facility relationship determination module mainly include three types of containing relationship, adjacent relationship, and connected relationship.

4. The community-oriented low-altitude aircraft path planning and capacity assessment system according to claim 1 or 2, characterized in that, The hierarchical airspace segmentation module uses a cubic grid with a grid edge length of 0.5-3 meters in a building-dense area and combines at least octree subdivision, and uses a cubic grid with a grid edge length of 5-20 meters in an open area.

5. The community-oriented low-altitude aircraft path planning and capacity assessment system according to claim 1 or 2, characterized in that, The noise-velocity function The expression is: ; wherein, is the sound pressure at a distance of 10 meters from the center of the aircraft during flight, k is a model-related coefficient for calibrating the differences in aerodynamic noise of different models, D0 is a reference sound pressure at a distance of 10 meters from the center of the aircraft, v is the actual flight speed of the aircraft, and c is the local speed of sound. Noise - distance function The expression for the noise - distance function is: ; Noise at target point The expression for the noise at the target point is: ; wherein, is the distance of the target point from the center of the vehicle, is the theoretical correction term for converting the sound power level to the sound pressure level for spherical wave propagation, is the environmental attenuation.

6. The community-oriented low-altitude aircraft path planning and capacity assessment system of claim 5, wherein, The lateral safety distance perpendicular to the flight direction under the condition of meeting the noise constraint A circular safety area is formed with a radius of ; wherein, is the noise limit value of different areas, The safety distance along the flight direction , and the maximum value of the two is taken while considering the noise impact; wherein, is the minimum safety distance; is the real-time speed of the aircraft in the grid; is the reaction time; a is the maximum braking acceleration.

7. The community-oriented low-altitude aircraft path planning and capacity assessment system of claim 6, wherein, The dynamic obstacle marking module, when the aircraft is flying towards a building, extends along the flight direction with the building as the reference The grid within the range is marked as "no pass"; when the aircraft path needs to pass between two buildings, extend vertically on both sides of the path direction with the edge line of the two buildings as the reference The range; identify the connected region by a clustering algorithm, and if the passable width of the connected region is less than , mark the connected region as an "obstacle section".

8. The community-oriented low-altitude aircraft path planning and capacity assessment system of claim 7, wherein, The total cost function of the improved A* algorithm is f(n) = g(n) + h(n); wherein g(n) is an actual cumulative cost from a starting point to a current node n, and h(n) is a heuristic estimated cost from the current node n to an end point. g(n) = ω1·L(n) + ω2·T(n) + ω3·Np(n) + ω4·Sp(n) + ω5·Fp(n); wherein, L(n) is the basic cost, T(n) is the time cost, Np(n) is the noise cost, Sp(n) is the total safety cost, and Fp(n) is the cross-border cost; the parameters ω1, ω2, ω3, ω4, ω5 correspond to the weights of the indexes L(n), T(n), Np(n), Sp(n), Fp(n) respectively, and are used to balance and integrate the index costs that need to be considered in multi-objective conflict; Heuristic estimation cost h(n) = α·three-dimensional Manhattan distance + β·building proximity penalty + γ·function adaptability; wherein, α, β, γ are weight coefficients; Building proximity penalty: if the distance from node n to the nearest building is < D ⊥safe then add 50; otherwise 0; The function adaptability is used to depict the consistency degree between the function attribute of the space or channel where the candidate node is located and the current flight task type: the node is reduced by 20 in the exclusive airspace, the node is reduced by 10 near the exclusive channel, and when the candidate node enters the exclusive airspace / dedicated channel of the non-task type, it is forbidden or added by 200.

9. The community-oriented low-altitude aircraft path planning and capacity assessment system of claim 8, wherein, The bottleneck identification and traffic capacity assessment module extracts bottleneck area parameters including the closest horizontal distance D between the outer contours of the two buildings. clear Bottleneck length L bottleneck Bottleneck area available height H bottleneck Wherein, the bottleneck section length L bottleneck The intersection of the main traffic path obtained from A* planning and the building outline is determined; Maximum allowed speed v in bottleneck region allow = min { v noise ,v ⊥ ,v brake}, where: v noise is the maximum speed resulting from noise limit constraints; v ⊥ is the maximum speed resulting from lateral safety distance constraints; v brake is the maximum speed resulting from forward braking safety constraints.

10. The community-oriented low-altitude aircraft path planning and capacity assessment system of claim 9, wherein, After determining the maximum allowable speed of the bottleneck area, the bottleneck identification and traffic capacity evaluation module abstracts the bottleneck area as a "single-channel cross section" low-altitude channel, calculates the single-aircraft passing time through the bottleneck section and the minimum time interval of adjacent aircrafts, and determines the single-channel traffic capacity; then, based on the number of parallel channels, the total traffic capacity of the bottleneck area is calculated.

Images