Multi-level low-altitude air route network construction method in complex urban environment

Abstract

The invention provides a multi-level low-altitude air route network construction method in a complex urban environment, and relates to the technical field of low-altitude air route construction, and the method comprises the steps: building a low-altitude flight rule base; performing three-dimensional discretization on the urban space, and dividing the airspace into three-dimensional grids; designing a risk assessment function R; dividing a plurality of height layers, and distributing the height layers for different types of aircrafts; constructing a multi-stage air route network; constructing an airspace three-dimensional airway model; based on the airspace three-dimensional air route model, adopting a hybrid iterative algorithm to optimize the air route planning of each height layer; performing path smoothing on the discrete path points generated by the hybrid iterative algorithm by adopting a B spline curve, and performing flight performance verification on the smoothed path to obtain a continuous flight path; and generating a low-altitude air route network according to the continuous flight path. According to the method, the defects in the aspects of complex urban environment processing, multi-source data integration, multi-level airspace division, path optimization and the like in the prior art can be overcome, so that safe, efficient and flexible low-altitude flight management is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of low-altitude route construction, and in particular to a method for constructing a multi-level low-altitude route network in a complex urban environment. Background Art

[0002] With the rapid development of drone technology, low-altitude aircraft are increasingly used in urban environments. However, the complex urban environment brings many challenges to low-altitude flight, including high-rise buildings, dense population, complex weather conditions, and communication signal interference. In order to ensure the safety and efficiency of low-altitude flight, establishing a complete low-altitude route network has become a hot topic of current research. At present, some researchers have proposed a public route network planning method based on obstacle Voronoi diagram and A* algorithm, which can solve the path planning problem in complex environments to a certain extent.

[0003] However, with the popularization of 5G technology and the complexity of low-altitude public routes, ensuring coverage has become a new challenge. From the perspective of time and space, existing route network planning methods are insufficient in dealing with large-scale 5G base stations and complex low-altitude public routes. In addition, the complexity of the urban surface environment also brings difficulties to route network planning. Due to the lack of geographic information and low-altitude route support, the current UAV traffic management system (UTM) faces great challenges in dealing with complex urban environments.

[0004] To solve these problems, some researchers have proposed a method to iteratively construct a route network for low-altitude urban drones. This method attempts to optimize the route network through multiple iterations to adapt to the complex urban environment. However, this method still has some limitations, such as the need to deal with dynamically changing environments and the needs of multiple types of aircraft.

[0005] In summary, existing technologies still have technical defects when dealing with low-altitude route network planning in complex urban environments, mainly manifested in the difficulty in ensuring full coverage, lack of sufficient consideration of the complex surface environment, and deficiencies in dealing with the needs of various types of aircraft and dynamic environmental changes. These problems limit the practical application and effectiveness of low-altitude route networks in complex urban environments. Summary of the invention

[0006] In view of this, the present invention proposes a multi-level low-altitude route network construction method in a complex urban environment to address the shortcomings of the existing technology in dealing with complex urban environments, multi-source data integration, multi-level airspace division and path optimization, thereby achieving safe, efficient and flexible low-altitude flight management.

[0007] The technical solution of the present invention is implemented as follows: The present invention provides a method for constructing a multi-level low-altitude airway network in a complex urban environment, comprising:

[0008] S1 Establish a low-altitude flight rules library;

[0009] S2 discretizes the urban space in three dimensions based on an adaptive octree structure, dividing the airspace into non-uniform three-dimensional grids;

[0010] S3 integrates geographic information system data, real-time meteorological data, population density data, and communication signal strength to form multi-source data; based on multi-source data and low-altitude flight rule library, a risk assessment function R is designed;

[0011] S4 divides the airspace from the ground to 1,000 meters into multiple altitude layers based on the low-altitude flight rule library, and assigns altitude layers to different types of aircraft;

[0012] S5 builds a multi-level route network, including the first-level route network, the second-level route network, the third-level route network, the fourth-level route network, the fifth-level route network and the sixth-level route network;

[0013] S6 builds a three-dimensional airspace route model based on the sixth-level route network in a three-dimensional grid environment, including route network model, traffic node model, take-off and landing point channel model and terminal route model;

[0014] S7 is based on the three-dimensional airspace route model and uses a hybrid iterative algorithm to optimize the route planning at each altitude level;

[0015] S8 uses B-spline curves to smooth the discrete path points generated by the hybrid iterative algorithm, verifies the flight performance of the smoothed path, and obtains a continuous flyable path;

[0016] S9 generates a low-altitude route network based on a continuous flyable path, including horizontal segments, vertical segments, intersections and terminal routes.

[0017] Based on the above solution, preferably, the low-altitude flight rule library includes:

[0018] Flight altitude restrictions: stipulate the minimum altitude for flying in a specific area;

[0019] Route regulations: specify the routes of aircraft within a specific area;

[0020] Flight speed limit: specifies the maximum flight speed of an aircraft in a specific area;

[0021] Air traffic control: Establish an air traffic control system to regulate low-altitude flying traffic order;

[0022] Emergency handling: Develop procedures for dealing with emergencies;

[0023] Violation Penalties: Penalties will be imposed on aircraft that violate traffic regulations.

[0024] Based on the above solution, preferably, in step S4, the altitude layer division includes:

[0025] 0-20m, near-ground airspace; 20-50m, isolated airspace for consumer-grade aircraft; 50-120m, suitable airspace for consumer-grade aircraft; 120-150m, isolated airspace for light and small aircraft; 150-200m, suitable airspace for light and small aircraft; 200-230m, isolated airspace for medium-sized aircraft; 230-300m, suitable airspace for medium-sized aircraft; 300-400m, isolated airspace for large aircraft; 400-600m, suitable airspace for large aircraft; 600-700m, isolated airspace for ultra-high-speed manned aircraft; 700-1000m, ultra-high-speed manned aircraft suitable airspace.

[0026] On the basis of the above scheme, preferably, in step S5, the first-level route network is an initial route network generated based on the ground route network; the second-level route network is a route network of iterative positively constrained geographical elements based on the scope of the first-level route network; the third-level route network is a route network of iterative negatively constrained geographical elements based on the second-level route network; the fourth-level route network is a dynamic route network based on the third-level route network for iterative aircraft traffic needs; the fifth-level route network is a route network that is corrected through iterative simulation flight tests on the fourth-level route network; and the sixth-level route network is a route network that is iteratively tested through actual flight tests based on the fifth-level route network.

[0027] Based on the above solution, preferably, the route network model includes:

[0028] Construct an elliptical cylindrical fixed hierarchical route pipeline to form a trunk route network, including flight path width, altitude and buffer zone;

[0029] The traffic node model includes:

[0030] The route network is abstracted as a collection of traffic nodes and flight routes. The traffic nodes are constructed as a ring-shaped spiral interchange. Multi-layer circular flight ramps are set within the intersection range, and the same-layer variable-speed channels are set outside the ramps.

[0031] The take-off and landing point channel model includes:

[0032] A conical space structure is used to connect the ground take-off and landing points with the air route network;

[0033] The terminal route model includes:

[0034] Connect the trunk route network with ground take-off and landing facilities, forced landing strips and related support facilities;

[0035] Among them, the terminal route model is connected with the main route network through a circular spiral interchange, which is an intersection where only right turns are allowed.

[0036] Based on the above solution, preferably, step S7 includes:

[0037] S71 calculates a priority value according to the mission information of the aircraft, assigns a priority to the aircraft based on the priority value, and classifies the aircraft into two categories of high priority and medium and low priority according to the priority;

[0038] S72 uses the first optimization algorithm to perform route planning for the high priority aircraft;

[0039] S73 uses the second optimization algorithm to plan routes for medium and low priority aircraft;

[0040] In the process of route planning, the constraints of the three-dimensional airspace route model are considered, including:

[0041] The main flight paths are searched in the route network model, following a fixed hierarchical pipe structure in the shape of an elliptical cylinder;

[0042] The traffic node model is used to realize the conversion between different altitude layers and routes, and a circular spiral interchange structure is adopted;

[0043] Plan the take-off and landing paths through the take-off and landing point path model, using a conical space structure;

[0044] Use the terminal route model to complete precise navigation in the mission area, connecting the main route network with the specific mission location.

[0045] Based on the above solution, preferably, step S72 includes:

[0046] S721 sorts the high-priority aircraft in descending order of priority;

[0047] S722 plans a path for each high-priority aircraft in order of priority, including:

[0048] Use a global path planning algorithm to find the fastest path from the start point to the end point in the three-dimensional airspace route model, while considering the flight time and risk assessment function R;

[0049] Considering the trajectory and time schedule of the planned high-priority aircraft, evaluating the potential conflict between the path of the current high-priority aircraft and the planned high-priority aircraft;

[0050] If there is a risk of conflict, adjust the flight plan of the current high-priority aircraft, including time adjustment, altitude adjustment, and path replanning;

[0051] Under the premise of satisfying all constraints, select the optimal path with the shortest expected arrival time;

[0052] S723 records the planned path and schedule of the current high-priority aircraft, and updates the trajectory information of the high-priority aircraft;

[0053] S724 repeats steps S722-723 until the path planning of all high priority aircraft is completed.

[0054] Based on the above solution, preferably, the path optimization objective function of the first optimization algorithm is as follows:

[0055]

[0056] R(x,y,z,t)=ω1·O(x,y,z)+ω2·P(x,y,z,t)+ω3·W(x,y,z,t)+ω4·T(x,y,z,t)

[0057] Where J1 is the total path cost, E1(t) is the energy consumption function of the high priority aircraft, λ is the risk weight factor, t0 and t f are the start and end time, (x(t), y(t), z(t)) is the spatial position of the aircraft at time t, R(x,y,z,t) represents the comprehensive risk value at the spatial position (x,y,z) and time t, O(x,y,z) is the obstacle risk function, P(x,y,z,t) is the population density risk function, W(x,y,z,t) is the meteorological risk function, T(x,y,z,t) is the communication risk function, ω1, ω2, ω3, ω4 are dynamically adjusted weight coefficients.

[0058] Based on the above solution, preferably, step S73 includes:

[0059] S731 builds a multi-objective path planning model based on the mission requirements, starting points and destinations of all medium and low priority aircraft, with optimization objectives including flight distance, energy consumption and route congestion;

[0060] S732 considers the constraints of the three-dimensional airspace route model, as well as the paths and time scheduling of planned high-priority aircraft, and establishes collision avoidance constraints and airspace accessibility constraints;

[0061] S733 performs iterative path optimization for medium and low priority aircraft, including:

[0062] The flight path is represented as a chromosome code based on the route node sequence; the initial population is generated based on the shortest path algorithm;

[0063] A multi-objective optimization function is used as the fitness function, taking into account the flight distance, energy consumption and route congestion;

[0064] An elite retention strategy is used for selection, and a partial mapping crossover method is used, combined with adaptive mutation probability for optimization;

[0065] After the iteration reaches the preset conditions, S734 outputs the optimal path sequence and completes the path planning of medium and low priority aircraft.

[0066] Based on the above solution, preferably, the multi-objective optimization function in the second optimization algorithm is as follows:

[0067] J2=w1·D+w2·E2+w3·C

[0068]

[0069] Where J2 is the comprehensive cost function, D is the flight distance, dist(n i ,n i+1 ) represents node n i and n i+1 The spatial distance between them, E2 is the energy consumption, e(n i ,n i+1 ) is the aircraft from node n i to n i+1 The energy consumed, C is the route congestion, γ(n i ) is node n i The route congestion evaluation function at is, N is the number of nodes, w1, w2, w3 are weight coefficients;

[0070] Route congestion evaluation function γ(n i ) is calculated as follows:

[0071]

[0072] In the formula, Q(n i ) is node n i The number of aircraft passing per unit time, Q max For node n i The maximum throughput capacity, S(n i ) is node n i The average size of the aircraft at S max is the maximum size of the aircraft, U(n i ) is node n i The cumulative priority penalty at max is the maximum value of the priority penalty.

[0073] Compared with the prior art, the present invention has the following beneficial effects:

[0074] (1) By comprehensively considering spatial discretization, multi-source data integration, multi-level airspace division, and path optimization, safe, efficient, and flexible low-altitude flight management is achieved. This method effectively solves the shortcomings of existing technologies in dealing with complex urban environments, ensuring comprehensive coverage, and adapting to environmental diversity and dynamics, and significantly improves the practical application effect of low-altitude route networks in complex urban environments;

[0075] (2) By integrating multiple data sources such as geographic information system data, real-time meteorological data, population density data, and communication signal strength, and combining them with the low-altitude flight rule library, a risk assessment function is designed. This comprehensive approach improves the accuracy and comprehensiveness of risk assessment;

[0076] (3) A priority-based hybrid iterative algorithm is proposed to plan routes for high-priority, medium-priority and low-priority aircraft respectively. This differentiated processing method effectively balances the needs of different types of aircraft, improves the utilization efficiency of the overall route network, and ensures the timely completion of high-priority tasks;

[0077] (4) The first optimization algorithm adopted realizes the optimal path planning of high-priority aircraft by considering energy consumption and comprehensive risk value. This method not only ensures the timely completion of high-priority tasks, but also minimizes energy consumption while ensuring flight safety, thereby improving the overall efficiency of high-priority flight tasks; (5) The dynamic adjustment weight coefficient mechanism introduced in the first optimization algorithm enables the system to flexibly adjust the importance of obstacle risk, population density risk, meteorological risk and communication risk according to real-time conditions. This adaptive mechanism significantly improves the flexibility and adaptability of route planning, and can better cope with the complex and changing urban flight environment;

[0078] (6) The second optimization algorithm uses a multi-objective optimization function, comprehensively considering flight distance, energy consumption, and route congestion, to achieve group optimization of medium and low priority aircraft. This method not only balances the performance requirements of a single aircraft, but also considers the load balancing of the entire route network;

[0079] (7) The congestion penalty mechanism and route congestion evaluation function introduced in the second optimization algorithm enable the system to dynamically adjust the flight path to avoid congested areas. This mechanism effectively reduces the occurrence of route congestion, improves the overall traffic efficiency of medium and low priority aircraft, and also reserves more route resources for high priority aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0081] Figure 1 Flow chart of the method of the present invention

[0082] Figure 2 It is a schematic diagram of the four stages of traffic prediction of the present invention;

[0083] Figure 3 It is a schematic diagram of traffic zone division of the present invention;

[0084] Figure 4 It is a traffic distribution schematic diagram of the present invention;

[0085] Figure 5 It is a schematic diagram of the route network model of the present invention;

[0086] Figure 6 It is a schematic diagram of traffic node coding of the present invention;

[0087] Figure 7 A schematic diagram of a traffic node model of the present invention;

[0088] Figure 8 It is a schematic diagram of the take-off and landing point channel model of the present invention;

[0089] Fig. 9 It is a schematic diagram of the terminal route model of the present invention;

[0090] Fig.10 It is a schematic diagram of the avoidance strategy of the present invention. DETAILED DESCRIPTION

[0091] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0092] like Figure 1 As shown, the present invention provides a method for constructing a multi-level low-altitude route network in a complex urban environment, comprising:

[0093] S1 Establish a low-altitude flight rules library;

[0094] S2 discretizes the urban space in three dimensions based on an adaptive octree structure, dividing the airspace into non-uniform three-dimensional grids;

[0095] S3 integrates geographic information system data, real-time meteorological data, population density data, and communication signal strength to form multi-source data; based on multi-source data and low-altitude flight rule library, a risk assessment function R is designed;

[0096] S4 divides the airspace from the ground to 1,000 meters into multiple altitude layers based on the low-altitude flight rule library, and assigns altitude layers to different types of aircraft;

[0097] S5 builds a multi-level route network, including the first-level route network, the second-level route network, the third-level route network, the fourth-level route network, the fifth-level route network and the sixth-level route network;

[0098] S6 constructs a three-dimensional airspace route model in a three-dimensional grid environment, including a route network model, a traffic node model, a take-off and landing point channel model, and a terminal route model;

[0099] S7 is based on the three-dimensional airspace route model and uses a hybrid iterative algorithm to optimize the route planning at each altitude level;

[0100] S8 uses B-spline curves to smooth the discrete path points generated by the hybrid iterative algorithm, verifies the flight performance of the smoothed path, and obtains a continuous flyable path;

[0101] S9 generates a low-altitude route network based on a continuous flyable path, including horizontal segments, vertical segments, intersections and terminal routes.

[0102] Specifically, in one embodiment of the present invention, the low-altitude flight rule library includes: flight altitude restrictions: specifying the minimum altitude for flying within a specific area; route regulations: specifying the route of an aircraft within a specific area; flight speed restrictions: specifying the maximum flight speed of an aircraft within a specific area; air traffic control: establishing an air traffic control system to supervise the traffic order of low-altitude flights; emergency response: formulating procedures for dealing with emergencies; and penalties for violations: imposing penalties on aircraft that violate traffic rules.

[0103] The low-altitude flight rule library is the basis for the construction of the route network. It is used to determine the altitude layer division standards of the airspace and the flight altitude restrictions of different types of aircraft, guide the design of route networks, traffic nodes, take-off and landing routes, and terminal routes, ensure compliance with aviation regulations and safety requirements, and provide specifications and restrictions for path planning, conflict resolution, and avoidance strategies to ensure that aircraft flights comply with regulations.

[0104] Specifically, in one embodiment of the present invention, an adaptive octree structure is used to discretize the urban space in three dimensions. The octree is a tree-shaped data structure for representing three-dimensional space. Each node represents a cubic space, which can be divided into eight sub-cubes. The three-dimensional discretization process is as follows: a) First, the entire urban space is regarded as a large cube. b) According to complex structures such as obstacles and buildings in the space, the cube is recursively divided into smaller sub-cubes. c) In areas with high complexity (such as high-rise buildings), more detailed segmentation is performed. d) In open areas (such as parks and squares), larger cubes are maintained. e) The segmentation is stopped when the preset minimum cube size or space complexity threshold is reached. f) Pointers or indexes are used to improve the access efficiency of the tree structure. Due to the use of an adaptive method, the final three-dimensional grid is uneven. That is, complex areas have smaller and denser grid cells, while simple areas have larger and sparser grid cells.

[0105] Specifically, in one embodiment of the present invention, step S3 includes:

[0106] Acquire multi-source data, including geographic information system data, real-time weather data, population density data, and communication signal strength. Geographic Information System (GIS) data: Use spatial databases (such as PostGIS) to store and manage urban three-dimensional spatial information. Real-time weather data: Obtain weather information in real time through API interfaces (such as OpenWeatherMap API). Population density data: Generate dynamic population density distribution using demographic databases and spatial interpolation algorithms (such as Kriging). Communication signal strength: Collect data in real time through distributed sensor networks and signal strength measurement devices.

[0107] The min-max normalization method is used to unify data of different scales into the interval [0,1]. For time-varying data, the sliding window technology is used for real-time standardization. The Kalman filter algorithm is used to fuse multi-source data to improve the accuracy and reliability of the data. The spatiotemporal database is used to store and manage the fused multidimensional data.

[0108] Design a mathematical model for risk assessment function R:

[0109] R(x,y,z,t)=ω1·O(x,y,z)+ω2·P(x,y,z,t)+ω3·W(x,y,z,t)+ω4

[0110] T(x,y,z,t)

[0111] R(x,y,z,t) represents the comprehensive risk value at the spatial position (x,y,z) and time t, O(x,y,z) is the obstacle risk function, P(x,y,z,t) is the population density risk function, W(x,y,z,t) is the meteorological risk function, T(x,y,z,t) is the communication risk function, ω1, ω2, ω3, and ω4 are dynamically adjusted weight coefficients.

[0112] Among them, the calculation method of each sub-function is as follows: Obstacle risk function O(x,y,z): Use R-tree spatial indexing technology to quickly retrieve obstacles around a given coordinate (x,y,z). Calculate the distance between the aircraft and the nearest obstacle and map it to the risk value using an inverse proportional function. Population density risk function P(x,y,z,t): Use spatiotemporal interpolation algorithms (such as spatiotemporal kriging) to process dynamic population density data. Map the interpolation results to risk values. The higher the population density, the greater the risk value. Meteorological risk function W(x,y,z,t): Updated using numerical weather forecast models and real-time meteorological data. Considering factors such as wind speed and visibility, the comprehensive meteorological risk is calculated using a fuzzy logic system. Communication risk function T(x,y,z,t): Calculate the signal strength at a given location based on a signal propagation model and measured data. Reversely map the signal strength to a risk value. The weaker the signal, the higher the risk. Use machine learning methods to dynamically adjust ω1,ω2,ω3,ω4 based on historical data and current conditions. Design a feedback mechanism to continuously optimize the weight coefficients based on actual flight data.

[0113] Computational optimization of risk assessment functions: Use GPU accelerated computing, such as CUDA or OpenCL, to improve the efficiency of large-scale spatial risk assessment. Implement spatial partitioning and parallel computing strategies to improve the real-time performance of risk assessment.

[0114] Integration with low-altitude flight rule base: Design a rule engine to convert the regulations in the low-altitude flight rule base into constraints. During the risk assessment process, dynamically check whether the flight rules are violated and adjust the risk value appropriately.

[0115] Specifically, in one embodiment of the present invention, the airspace spatial height segmentation is established on the vertical plane of the airspace, and is extended upward according to the flight altitude layer of the aircraft to establish the vertical height division. The low-altitude airspace (0-1000m) is subdivided into eleven altitude layers according to the altitude for different types of aircraft to fly, specifically:

[0116] 0-20m, near-ground airspace; 20-50m, isolated airspace for consumer-grade aircraft; 50-120m, suitable airspace for consumer-grade aircraft; 120-150m, isolated airspace for light and small aircraft; 150-200m, suitable airspace for light and small aircraft; 200-230m, isolated airspace for medium-sized aircraft; 230-300m, suitable airspace for medium-sized aircraft; 300-400m, isolated airspace for large aircraft; 400-600m, suitable airspace for large aircraft; 600-700m, isolated airspace for ultra-high-speed manned aircraft; 700-1000m, ultra-high-speed manned aircraft suitable airspace.

[0117] It should be noted that the airspace division of the present invention is based on true height, that is, the vertical height of the aircraft from a certain reference horizontal plane. It is a relative height division, not an absolute height. Specifically, it can be based on sea level or the ground. When actually dividing, it is determined according to the specific situation.

[0118] A hierarchical data structure (such as a tree structure) is used to represent these 11 altitude layers. Each altitude layer node contains the following information: altitude range (lower and upper limits), applicable aircraft type, whether it is isolated airspace, and current airspace usage status (such as the number of real-time aircraft, congestion, etc.).

[0119] Aircraft classification and altitude layer allocation algorithm:

[0120] a) Based on the physical characteristics of the aircraft (such as size, weight, flight performance) and mission requirements, the aircraft are classified into consumer, light small, medium, large and manned aircraft.

[0121] b) Design an altitude allocation algorithm: Input: aircraft type, mission requirements, current airspace usage status; Output: allocated altitude level; Algorithm flow: Determine the range of available altitude levels based on aircraft type; Check whether the mission requirements conflict with certain altitude levels (such as near-ground missions); Evaluate the current usage status of each altitude level to avoid allocation to congested altitude levels; If there are multiple optional altitude levels, use a heuristic algorithm (such as considering energy consumption minimization) to select the optimal altitude level; Return the allocation result.

[0122] Specifically, in one embodiment of the present invention, in step S5, the first-level route network is an initial route network generated based on the ground route network; the second-level route network is a route network that iterates positively constrained geographical elements based on the scope of the first-level route network; the third-level route network is a route network that iterates negatively constrained geographical elements based on the second-level route network; the fourth-level route network is a dynamic route network that iterates aircraft traffic needs based on the third-level route network; the fifth-level route network is a route network that is corrected after iterative simulation flight tests are performed on the fourth-level route network; and the sixth-level route network is a route network that iterates actual flight tests based on the fifth-level route network.

[0123] In this embodiment, before constructing a multi-level route network, an environment map needs to be constructed first. The specific process is as follows:

[0124] The flight environment of drones is highly complex. To ensure that the route can effectively avoid obstacles and fly safely, it is necessary to first identify and extract the impact boundaries of various geographic constraint elements (including safe clearance range) and build a geographic fence database for drone low-altitude flight environment, that is, an environmental map. The scope of geographic constraint elements refers to the "Urban Land Classification and Planning and Construction Land Standards" and the "Land Use Status Classification".

[0125] 1. Positive Constraint Geographical Elements

[0126] Parks and green spaces, water areas, roads, squares and other service facilities are positively constrained geographical elements that can be used by drones for low-altitude flight.

[0127] 2. Negatively constrained geographic elements

[0128] Areas prone to severe weather, areas with weak base station signals (communication blind spots), special restricted areas (party and government agencies at the municipal level and above, railway stations, passenger stations, port terminals, airports, high-rise buildings, schools, hospitals, power lines, overpasses, etc.), human factors and community activities (crowded areas), etc. are negative constraint geographical elements that drones need to avoid during flight.

[0129] The construction contents of the specific environment map are shown in the following table:

[0130]

[0131]

[0132] After constructing the environment map, this embodiment sequentially constructs a multi-level route network, as follows:

[0133] Level 1 route network: The regional low-altitude routes are closely related to the ground route network, so the initial route network is generated based on the ground route network. When determining the initial route network, the data of the third national land survey (hereinafter referred to as the "third survey") can be used as the basis, and the flight space corresponding to the map spots with the attribute codes of 1003, 1004 and 1006 in the third survey can be selected as the scope of the initial route network.

[0134] Level 2 route network: Based on the scope of the initial route network, the flight area environment is buffered 100m outward along the normal direction of the road centerline to determine whether the positive constrained geographical elements are connected to the level 1 routes. Those that are determined to be "connected" are included in the route network scope to generate the level 2 route network.

[0135] Level 3 route network: After generating the level 2 route network based on positive constraint elements, the drone must avoid sensitive geographical elements such as buildings, power lines, and overpasses during flight. By detecting potential conflicting sections based on a set of negative constraint elements, the conflicting sections are replanned to construct a level 3 route network.

[0136] Level 4 route network: The “four stages” of road traffic demand forecasting include traffic generation and attraction, traffic distribution, traffic mode division, and traffic allocation, such as Figure 2 As shown in the figure, drawing on the research ideas of the "four-stage" method for traffic demand prediction, based on the third-level route network, according to the proposed four-stage method for drone transportation demand prediction, combined with the actual drone traffic demand, traffic allocation is performed on the third-level route network (using the shortest path traffic allocation method) to form a dynamic low-altitude route network and construct a fourth-level route network. When constructing this level of route network, the schematic diagram of traffic zone division is shown in the figure below. Figure 3 As shown in the figure, the traffic distribution diagram is as follows Figure 4 shown.

[0137] Level 5 route network: In order to verify the safety and scientificity of the route, it is necessary to conduct simulated flight tests on the route construction results. Feedback the simulation results to the level 4 route network and correct the flight route to generate the level 5 route network. During the simulated flight, the current position of the drone is detected and calculated in real time. If the real-time position of the drone is within the route, the route is reasonable; otherwise, the level 4 route needs to be adjusted and corrected. Combined with the simulation results and repeated debugging, the parameter thresholds such as the route width, the minimum turning radius of the drone, and the pitch angle of the drone are finally determined.

[0138] Level 6 route network: In order to ensure the absolute safety of UAV flight activities, it is necessary to verify the authenticity and accuracy of the flight area environment through actual flight, and focus on monitoring the accuracy of communication link parameters, local climate environment modeling accuracy, and various geographic information such as urban buildings, road traffic, municipal infrastructure, terrain, power grid towers, etc., in order to generate a level 6 route network.

[0139] Specifically, in one embodiment of the present invention, the route network model includes: constructing an elliptical cylindrical fixed layered route pipeline to form a trunk route network, including flight route width, height and buffer zone; the traffic node model includes: abstracting the route network as a collection of traffic nodes and flight routes, constructing the traffic nodes in the form of a ring-shaped spiral interchange, setting up multi-layer circular flight ramps within the intersection range, and setting up the same-layer variable speed channels outside the ramps; the take-off and landing point channel model includes: using a conical space structure to connect the ground take-off and landing points with the air route network; the terminal route model includes: connecting the trunk route network with ground take-off and landing facilities, forced landing strips and related support facilities; wherein the terminal route model is connected to the trunk route network through a ring-shaped spiral interchange, and the interchange is an intersection where only right turns are allowed.

[0140] Let's take a specific example to illustrate:

[0141] (1) Route network model

[0142] Aircraft perform flight missions at corresponding flight altitude layers according to their types. The flight route is an elliptical fixed layered pipe structure composed of a starting point, an end point, and a flight altitude. During flight, the aircraft's flight altitude, direction, and speed need to be restricted and required to reduce the complexity of the flight mode within the channel structure and improve flight safety. The width of the flight channel is determined based on the aircraft type, degree of automation, and other performance characteristics. The aircraft can fly within it at a minimum safe distance. At the same time, a buffer zone of a certain width must be set around the channel. No other channels or obstacles are allowed in the buffer zone, and the buffer zones of different channels can overlap. For example Figure 5 As shown, Figure 5 A multi-layered route structure is shown, with each layer representing a specific flight altitude. This layered design allows aircraft of different types or missions to operate at their own altitude layers, effectively reducing the risk of vertical conflicts. The flight path of each aircraft is designed to be cylindrical, a shape that is more in line with the movement characteristics of the aircraft than a simple rectangular channel, especially when turning. The flight paths of all aircraft together constitute the flight path of this layer. In order to simplify management and recognition by the automatic system, the flight path is designed to have a rectangular channel shape. The cross-section is rectangular, which not only provides ample operating space but also ensures clarity of the boundaries. Figure 5 W is the flight channel width, H is the flight channel height, R is the aircraft safety clearance, Di is the lateral clearance, and Li is the track safety clearance. Although the model stipulates a fixed route structure, it can adapt to the needs of different types of aircraft by adjusting W (channel width) and H (channel height). A buffer zone is set around each channel, and other channels or obstacles are not allowed to enter, which further increases the safety margin. The overlapping design of the buffer zone ensures safety and improves space utilization efficiency.

[0143] (2) Traffic node model

[0144] The route network is abstracted into a collection of traffic nodes and flight routes. For example, a 3×3 route network can be abstracted into a network of 9 traffic nodes and 24 one-way routes. Figure 6 As shown in the figure, assuming that the aircraft flies according to the right-hand rule in the air, the route section is encoded in combination with the route heading:

[0145]

[0146]

[0147] The route network nodes are in the form of circular spiral interchanges, with multi-layered circular flight ramps set up within the intersection, and speed-changing channels (acceleration channels or deceleration channels) on the same level set up outside the ramps. When the aircraft passes through a traffic node, it must first rise or fall vertically to the ramp height, rise or fall to the ramp height in the target direction through the circular spiral interchange, and then rise or fall vertically to the original channel height through the speed-changing channel (acceleration channel or deceleration channel) of the ramp at that height.

[0148] like Figure 7 As shown, Figure 7 At least five different altitude layers are displayed, including the main route layer (layer ①) and multiple import and export layers (layers ②, ③, ④, and ⑤). The aircraft flying from west to east flies horizontally to the deceleration channel entrance of the route layer (layer ①), descends vertically to the deceleration channel entrance of the west import channel layer (layer ④), decelerates and flies horizontally for a certain distance to the interchange ramp entrance of the west import channel layer (layer ④), passes through the spiral interchange, enters the east exit channel layer (layer ②), accelerates horizontally for a certain distance to the acceleration channel entrance of this layer (layer ②), and descends vertically to the route layer (layer ①). Similarly, an aircraft flying from east to west will fly horizontally to the deceleration channel entrance of the route layer (layer ①), descend vertically to the deceleration channel entrance of the east entrance channel layer (layer ⑤), decelerate and fly horizontally for a certain distance to the interchange ramp entrance of the east entrance channel layer (layer ⑤), pass the spiral interchange, enter the west exit channel layer (layer ③), accelerate horizontally for a certain distance to the acceleration channel entrance of this layer (layer ③), and descend vertically to the route layer (layer ①). Figure 7 For illustration, assume that two aircraft A and B meet head-on in the east-west direction and need to pass through this intersection. 1. Aircraft A (west to east): When approaching the intersection at the route layer (layer ①), enter the deceleration channel. Descend vertically to the west entrance channel layer (layer ④). Continue to decelerate in the deceleration channel of layer ④. Enter the spiral interchange and rise to the east exit channel layer (layer ②). Accelerate in the acceleration channel of layer ②. Finally, descend vertically back to the route layer (layer ①) and continue to travel east. 2. Aircraft B (east to west): When approaching the intersection at the route layer (layer ①), enter the deceleration channel. Descend vertically to the east entrance channel layer (layer ⑤). Continue to decelerate in the deceleration channel of layer ⑤. Enter the spiral interchange and rise to the west exit channel layer (layer ③). Accelerate in the acceleration channel of layer ③. Finally, descend vertically back to the route layer (layer ①) and continue to travel west. Aircraft A and B are always at different altitudes throughout the process, completely avoiding the risk of collision. The spiral interchange design allows the aircraft to change direction and altitude smoothly, reducing sharp turns and sudden altitude changes. This design can easily accommodate traffic flows in more directions by simply adding corresponding entry and exit levels.

[0149] (3) Take-off and landing point channel model

[0150] like Figure 8As shown, considering that some manned aircraft must accelerate to a certain speed during takeoff to enter the corresponding layer for normal flight, their takeoff trajectory is not completely vertical. Therefore, the channel airspace of the take-off and landing points is set as a "conical" space for the aircraft to safely pass through the drone layer.

[0151] (4) Terminal route model

[0152] like Fig. 9 As shown in the figure, the terminal route is the last kilometer route connecting the urban route with the ground take-off and landing facilities, emergency landing strips and related support facilities, ensuring that the aircraft can get timely assistance or land safely nearby when a fault or abnormal situation occurs during the flight. Its design is closely related to the location of ground take-off and landing facilities, emergency landing strips and other facilities. The terminal route is connected to the urban route network through a circular spiral interchange. The interchange here is an intersection that only allows right turns. At this type of intersection, only the right turn route needs to pass through the spiral interchange.

[0153] Specifically, in one embodiment of the present invention, step S7 includes:

[0154] S71 calculates a priority value according to the mission information of the aircraft, assigns a priority to the aircraft based on the priority value, and classifies the aircraft into two categories of high priority and medium and low priority according to the priority;

[0155] Specifically, when calculating the priority value, the mission importance, urgency, and time sensitivity are taken into consideration, and the priority value of each aircraft is calculated in a weighted sum manner. According to the calculated priority value, an appropriate threshold is set. Those above the threshold are classified as high priority, and those below the threshold are classified as medium or low priority.

[0156] S72 uses the first optimization algorithm to perform route planning for the high priority aircraft;

[0157] S73 uses the second optimization algorithm to plan routes for medium and low priority aircraft;

[0158] Among them, in the process of route planning, the constraints of the three-dimensional route model of the airspace are taken into consideration, including: searching for the main flight path in the route network model, following the fixed hierarchical pipeline structure of the elliptical cylinder; using the traffic node model to realize the conversion between different altitude layers and routes, and adopting a ring-shaped spiral interchange structure; planning the path of the take-off and landing phases through the take-off and landing point channel model, using a conical space structure; using the terminal route model to complete precise navigation in the mission area, connecting the main route network with the specific mission location.

[0159] In this embodiment, step S72 includes:

[0160] S721 sorts the high-priority aircraft in descending order of priority;

[0161] S722 plans a path for each high-priority aircraft in order of priority, including: using a global path planning algorithm to find the fastest path from the starting point to the end point in the three-dimensional airspace route model, while considering the flight time and risk assessment function R; considering the trajectory and time scheduling of the planned high-priority aircraft, and evaluating the potential conflict between the path of the current high-priority aircraft and the planned high-priority aircraft; if there is a risk of conflict, adjusting the flight plan of the current high-priority aircraft, including time adjustment, altitude adjustment and path replanning; and selecting the optimized path with the shortest expected arrival time under the premise of satisfying all constraints;

[0162] S723 records the planned path and schedule of the current high-priority aircraft, and updates the trajectory information of the high-priority aircraft;

[0163] S724 repeats steps S722-723 until the path planning of all high priority aircraft is completed.

[0164] The path optimization objective function of the first optimization algorithm is as follows:

[0165]

[0166] R(x,y,z,t)=ω1·O(x,y,z)+ω2·P(x,y,z,t)+ω3·W(x,y,z,t)+ω4·T(x,y,z,t)

[0167] Where J1 is the total path cost, E1(t) is the energy consumption function of the high priority aircraft, λ is the risk weight factor, t0 and t f are the start and end time, (x(t), y(t), z(t)) is the spatial position of the aircraft at time t, R(x,y,z,t) represents the comprehensive risk value at the spatial position (x,y,z) and time t, O(x,y,z) is the obstacle risk function, P(x,y,z,t) is the population density risk function, W(x,y,z,t) is the meteorological risk function, T(x,y,z,t) is the communication risk function, ω1, ω2, ω3, ω4 are dynamically adjusted weight coefficients.

[0168] Specifically, the energy consumption function calculation formula is as follows:

[0169] E1(t)=p base (t)+p kinetic (t)+p potential (t)+p wind (t)+p payload (t)

[0170] In the formula, p base (t) is the basic power consumption, p base(t) = k b *m, which is proportional to the mass m of the aircraft, k b is the basic power coefficient; p kinetic (t) is the power required to overcome air resistance, p kinetic (t)=0.5*ρ(h)*C d *A*v 3 (t), ρ(h) is the air density at height h, C d is the drag coefficient, A is the frontal area of ​​the aircraft, v(t) is the flight speed at time t; p potential (t) is the power required to change the altitude, p potential (t) = m*g*dh / dt, g is the acceleration due to gravity, dh / dt is the vertical velocity, rising is positive and falling is negative; p wind (t) is the effect of wind on energy consumption, p wind (t) = k w *|v wind (t)|*cos(θ(t)), k w is the wind influence coefficient, v wind (t) is the wind speed at time t, θ(t) is the angle between the wind direction and the flight direction; p payload (t) is the additional power consumption caused by the load, p payload (t) = k p *m payload (t), k p is the load power factor, m payload (t) is the load mass at time t.

[0171] In this embodiment, the energy consumption function E1(t) is a continuous function of time, which represents the real-time energy consumption rate of the high-priority aircraft during flight. At each time point t, the energy consumption E1(t) of the aircraft and the comprehensive risk R of the current position (x, y, z) are considered, and the total cost of the entire flight process is calculated by integration, achieving a balance between safety and efficiency.

[0172] In this embodiment, step S73 includes:

[0173] S731 builds a multi-objective path planning model based on the mission requirements, starting points and destinations of all medium and low priority aircraft, with optimization objectives including flight distance, energy consumption and route congestion;

[0174] S732 considers the constraints of the three-dimensional airspace route model, as well as the paths and time scheduling of planned high-priority aircraft, and establishes collision avoidance constraints and airspace accessibility constraints;

[0175] Specifically, this embodiment is to prioritize the planning of high-priority aircraft paths. When planning medium- and low-priority aircraft paths, more constraints will be considered. The following is a specific embodiment description:

[0176] Collision avoidance constraints:

[0177] 1) Space-time separation constraint: for any two aircraft i and j, at any time t, the distance between them must be greater than the safety distance D safe :||p i (t)-p j (t)||≥D safe ; where p i (t) and p j (t) represent the position vectors of aircraft i and j at time t respectively.

[0178] 2) Velocity vector constraint: The direction of the relative velocity must be greater than a certain safety angle θ with the direction of the relative position vector. safe :(v i (t)-v j (t))·(p i (t)-p j (t))≤‖v i (t)-v j (t)‖·||p i (t)-p j (t)||·cos(θ safe ), where v i (t) and v j (t) represent the velocity vectors of aircraft i and j at time t respectively.

[0179] 3) Altitude separation constraint: For aircraft at different altitudes, the vertical separation must be greater than the minimum safe altitude H safe :|h i (t)-h j (t)|≥H safe , where h i (t) and h j (t) represent the flight altitudes of aircraft i and j at time t respectively.

[0180] Airspace accessibility constraints:

[0181] 1) Altitude constraint: The aircraft must be at its assigned altitude level H min and H max Flight between: H min ≤h(t)≤H max , where h(t) represents the flight altitude of the aircraft at time t.

[0182] 2) No-fly zone constraint: For any no-fly zone k, the aircraft cannot enter the area at any time t: ‖p(t)-c k ‖≥r k , where p(t) represents the position of the aircraft at time t, c k and r k represent the center and radius of the no-fly zone k respectively.

[0183] 3) Pathway constraint: The aircraft must fly within a predefined elliptical pathway: ((x(t)-x c (t)) 2 / a 2 )+((y(t)-y c (t)) 2 / b 2 )≤1, where (x(t),y(t)) represents the horizontal position of the aircraft at time t, (x c (t),y c (t)) represents the horizontal position of the route centerline at that moment, and a and b represent the lengths of the horizontal and vertical semi-axes of the ellipse, respectively.

[0184] 4) Turning radius constraint: The turning radius of the aircraft must not be less than the minimum safe turning radius R min :‖v(t)×a(t)‖ / ‖v(t)‖ 2 ≥1 / R min , where v(t) and a(t) represent the velocity and acceleration vector of the aircraft at time t, respectively.

[0185] 5) Communication coverage constraint: The aircraft must always remain within the communication coverage: ‖p(t)-p base ‖≤R comm , where p(t) represents the position of the aircraft at time t, p base represents the location of the communication base station, R comm Indicates the communication coverage radius.

[0186] S733 performs iterative path optimization for medium and low priority aircraft, including: representing the flight path as a chromosome code based on the route node sequence; generating the initial population based on the shortest path algorithm; using a multi-objective optimization function as the fitness function, taking into account the flight distance, energy consumption and route congestion; adopting an elite retention strategy for selection, using a partial mapping crossover method, and combining adaptive mutation probability for optimization.

[0187] Specifically, iterative optimization uses genetic algorithms for optimization, and enhances the algorithm's optimization ability through elite retention strategy, partial mapping crossover method, and adaptive mutation probability, as follows:

[0188] The elite retention strategy is a method to ensure that the best individuals are not lost during the evolution process. In each generation, a certain number of the best individuals (elites) are directly copied to the next generation. These elite individuals do not participate in crossover and mutation operations and maintain their original characteristics. The steps include: determining the elite ratio (usually 1-5% of the population size). At the end of each generation, all individuals are sorted according to the fitness function. The top-ranked elite individuals are directly copied to the next generation. The remaining positions are filled through conventional selection, crossover and mutation operations.

[0189] The Partial Mapping Crossover (PMX) method is a crossover operation designed specifically for permutation problems. Exchange a portion of genes between two parent chromosomes. Resolve conflicts that may arise after the crossover through mapping relationships to ensure that the generated offspring is still a valid permutation. The steps include: Randomly select two crossover points and determine the crossover region. Copy the crossover region of one parent directly to the same position of the offspring. Find the genes that have not been copied from the other parent. If the original positions of these genes are occupied, find new positions based on the mapping relationship. Fill the remaining genes into the vacancies of the offspring.

[0190] Adaptive mutation probability is a method of dynamically adjusting the mutation rate to balance global exploration and local development. The mutation probability is dynamically adjusted according to the evolutionary state of the population. When the population diversity decreases, the mutation probability is increased to promote exploration. When a good solution is found, the mutation probability is reduced for local optimization. The steps include: at the beginning of each generation, the average fitness and minimum fitness of the population are calculated. For each individual, its mutation probability is calculated according to its fitness value. The mutation operation is performed using the calculated mutation probability. The calculation formula for adaptive mutation probability is as follows:

[0191]

[0192] Among them, P m The mutation probability of the current individual; is the maximum mutation probability; is the minimum mutation probability; f is the fitness value of the current individual; f min is the minimum fitness value in the current population; f avg is the average fitness value of the current population.

[0193] After the iteration reaches the preset conditions, S734 outputs the optimal path sequence and completes the path planning of medium and low priority aircraft.

[0194] The multi-objective optimization function in the second optimization algorithm is as follows:

[0195] J2=w1·D+w2·E2+w3·C

[0196]

[0197]

[0198] Where J2 is the comprehensive cost function, D is the flight distance, dist(n i ,n i+1 ) represents node n i and n i+1 The spatial distance between them, E2 is the energy consumption, e(n i ,n i+1 ) is the aircraft from node n i to n i+1 The energy consumed, C is the route congestion, γ(n i ) is node n i The route congestion evaluation function at , N is the number of nodes, w1, w2, w3 are weight coefficients.

[0199] It should be noted that, in this embodiment, E2 is a discrete sum, and the sum of energy consumption between all adjacent nodes on the entire flight path is calculated. The comprehensive cost function J2 comprehensively considers three key factors: flight distance D, energy consumption E2, and route congestion C. By introducing weight coefficients w1, w2, and w3, the function can flexibly adjust the relative importance of these three factors to adapt to different mission requirements and environmental conditions. The flight distance D is calculated by accumulating the spatial distance between adjacent nodes, which directly affects the flight time and efficiency. The energy consumption E2 takes into account the energy consumption of the aircraft on the entire path, which helps to optimize battery use and extend flight time. The route congestion C helps avoid busy traffic areas and improve the efficiency and safety of the overall system by evaluating the congestion of each node on the path.

[0200] Route congestion evaluation function γ(n i ) is calculated as follows:

[0201]

[0202] In the formula, Q(n i ) is node n i The number of aircraft passing per unit time, Q max For node n i The maximum throughput capacity, S(n i ) is node n i The average size of the aircraft at S max is the maximum size of the aircraft, U(n i ) is node n i The cumulative priority penalty at max is the maximum value of the priority penalty.

[0203] In this embodiment, the route congestion evaluation function combines three key factors: the number of aircraft passing Q (n i ), the average size of the aircraft S(n i ) and the cumulative priority penalty U(n i ). By comparing and weighting these factors with their respective maximum values, the function is able to comprehensively assess the congestion level of a node. The number of aircraft passing reflects the traffic flow, the average size takes into account the space occupied by different types of aircraft, and the cumulative priority penalty introduces additional considerations for high-priority aircraft. This design can not only effectively identify and avoid congested areas, but also make way for high-priority aircraft when necessary, improving the flexibility and responsiveness of the entire system.

[0204] Specifically, in one embodiment of the present invention, the hybrid iterative algorithm generates discrete path points. These discrete points are interpolated using B-spline curves to generate smooth continuous curves. The advantage of B-spline curves is that they can maintain the overall shape of the path while providing a smooth transition. The smoothed path is subjected to flight performance verification to ensure that the path meets the dynamic constraints of the aircraft. The verification content includes: a) whether the turning radius meets the minimum turning radius requirement of the aircraft; b) whether the climb and descent rates are within the performance range of the aircraft; c) whether the speed change meets the acceleration limit of the aircraft; if a part that does not meet the flight performance requirements is found, local adjustments need to be made. After smoothing and verification, a continuous path that meets the performance requirements of the aircraft is obtained. This path ensures both smoothness and flyability.

[0205] Specifically, in one embodiment of the present invention, a complete low-altitude route network is generated based on a continuous flyable path, and its components include:

[0206] Horizontal segment: The straight or curved flight section of an aircraft within the same altitude layer, including trunk routes and branch routes.

[0207] Vertical segment: The portion of an aircraft's climb or descent between different altitude levels. It is used to connect routes at different altitude levels.

[0208] Intersection Interchange: Based on the traffic node model described above, a circular spiral interchange is adopted. A multi-layered spiral flight ramp is set up within the intersection. A variable speed channel on the same level is set outside the ramp. This design can effectively handle the intersection of routes in different directions and heights.

[0209] Terminal route: connects the main route network with ground take-off and landing facilities, forced landing strips and related support facilities. It adopts a conical space structure to connect the ground take-off and landing points with the air route network. It is connected with the main route network through a circular spiral interchange. This interchange is a right-turn-only intersection to simplify traffic flow and improve safety.

[0210] Specifically, in another embodiment of the present invention, when constructing a low-altitude route network, an air avoidance strategy can also be set, such as Fig.10 As shown:

[0211] There are three types of priority:

[0212] (1) Based on coordinated avoidance under cooperative communication surveillance, a conflict resolution strategy is established under real-time duplex communication conditions between aircraft with the same priority and equipment assembly level.

[0213] (2) Active avoidance based on established rules: low-priority or low-route-right aircraft proactively give way to high-priority aircraft under cooperative surveillance conditions.

[0214] (3) Autonomous sensing and avoidance under non-cooperative conditions: Under normal operating conditions, an aircraft can sense an unconnected “illegal” target or a malfunctioning and out-of-control aircraft based on its onboard sensing equipment and take active avoidance actions.

[0215] It should be noted that the purpose of the present invention to first construct a multi-level iterative route network is to gradually improve and optimize the route network, starting from the initial route based on the ground road network, and then consider the positive and negative constrained geographical elements, traffic demand prediction, simulation test and actual flight test results in turn, and finally form a 6th level route network that comprehensively considers various factors. This step-by-step iterative method can fully consider the complexity of the urban environment, including geographical constraints, traffic demand, safety and other aspects, to ensure that the final route network is both practical and safe. On this basis, hybrid iterative optimization is carried out to further improve the efficiency and safety of the route network. By dividing aircraft into high priority and medium and low priority, and using different optimization algorithms, the needs of different types of aircraft can be better balanced, while considering factors such as route congestion and energy consumption. This multi-level and multi-stage optimization method can comprehensively consider various complex factors of low-altitude flight, and finally construct a safe, efficient and flexible low-altitude route network, laying the foundation for the development of future urban air traffic.

[0216] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing a multi-level low-altitude air route network in a complex urban environment, characterized in that: include: S1 Establish a low-altitude flight rules library; S2 discretizes the urban space in three dimensions based on an adaptive octree structure, dividing the airspace into non-uniform three-dimensional grids; S3 integrates geographic information system data, real-time meteorological data, population density data, and communication signal strength to form multi-source data; based on multi-source data and low-altitude flight rule library, a risk assessment function R is designed; S4 divides the airspace from the ground to 1,000 meters into multiple altitude layers based on the low-altitude flight rule library, and assigns altitude layers to different types of aircraft; S5 builds a multi-level route network, including the first-level route network, the second-level route network, the third-level route network, the fourth-level route network, the fifth-level route network and the sixth-level route network; S6 builds a three-dimensional airspace route model based on the sixth-level route network in a three-dimensional grid environment, including route network model, traffic node model, take-off and landing point channel model and terminal route model; S7 is based on the three-dimensional airspace route model and uses a hybrid iterative algorithm to optimize the route planning at each altitude level; S8 uses B-spline curves to smooth the discrete path points generated by the hybrid iterative algorithm, verifies the flight performance of the smoothed path, and obtains a continuous flyable path; S9 generates a low-altitude route network based on a continuous flyable path, including horizontal segments, vertical segments, intersections and terminal routes.

2. The method for constructing a multi-level low-altitude air route network in a complex urban environment as claimed in claim 1, characterized in that: The low-altitude flight rules library includes: Flight altitude restrictions: stipulate the minimum altitude for flying in a specific area; Route regulations: specify the routes of aircraft within a specific area; Flight speed limit: specifies the maximum flight speed of an aircraft in a specific area; Air traffic control: Establish an air traffic control system to regulate low-altitude flying traffic order; Emergency handling: Develop procedures for dealing with emergencies; Violation Penalties: Penalties will be imposed on aircraft that violate traffic regulations.

3. The method for constructing a multi-level low-altitude air route network in a complex urban environment as claimed in claim 2, characterized in that: In step S4, the altitude layer division includes: 0-20m, near-ground airspace; 20-50m, isolated airspace for consumer-grade aircraft; 50-120m, suitable airspace for consumer-grade aircraft; 120-150m, isolated airspace for light and small aircraft; 150-200m, suitable airspace for light and small aircraft; 200-230m, isolated airspace for medium-sized aircraft; 230-300m, suitable airspace for medium-sized aircraft; 300-400m, isolated airspace for large aircraft; 400-600m, suitable airspace for large aircraft; 600-700m, isolated airspace for ultra-high-speed manned aircraft; 700-1000m, ultra-high-speed manned aircraft suitable airspace.

4. The method for constructing a multi-level low-altitude air route network in a complex urban environment as claimed in claim 1, characterized in that: In step S5, the first-level route network is an initial route network generated based on the ground road network; the second-level route network is a route network that iterates positively constrained geographical elements based on the scope of the first-level route network; the third-level route network is a route network that iterates negatively constrained geographical elements based on the second-level route network; the fourth-level route network is a dynamic route network that iterates aircraft traffic needs based on the third-level route network; the fifth-level route network is a route network that is corrected after iterative simulation flight tests are performed on the fourth-level route network; and the sixth-level route network is a route network that iterates actual flight tests based on the fifth-level route network.

5. The method for constructing a multi-level low-altitude air route network in a complex urban environment as claimed in claim 2, characterized in that: The route network model includes: Construct an elliptical cylindrical fixed hierarchical route pipeline to form a trunk route network, including flight path width, altitude and buffer zone; The traffic node model includes: The route network is abstracted as a collection of traffic nodes and flight routes. The traffic nodes are constructed as a ring-shaped spiral interchange. Multi-layer circular flight ramps are set within the intersection range, and variable-speed channels on the same level are set outside the ramps. The take-off and landing point channel model includes: A conical space structure is used to connect the ground take-off and landing points with the air route network; The terminal route model includes: Connect the trunk route network with ground take-off and landing facilities, forced landing strips and related support facilities; Among them, the terminal route model is connected with the main route network through a circular spiral interchange, which is an intersection where only right turns are allowed.

6. The method for constructing a multi-level low-altitude air route network in a complex urban environment as claimed in claim 5, characterized in that: Step S7 includes: S71 calculates a priority value according to the mission information of the aircraft, assigns a priority to the aircraft based on the priority value, and classifies the aircraft into two categories of high priority and medium and low priority according to the priority; S72 uses the first optimization algorithm to perform route planning for the high priority aircraft; S73 uses the second optimization algorithm to plan routes for medium and low priority aircraft; In the process of route planning, the constraints of the three-dimensional airspace route model are considered, including: The main flight paths are searched in the route network model, following a fixed hierarchical pipe structure in the shape of an elliptical cylinder; The traffic node model is used to realize the conversion between different altitude layers and routes, and a circular spiral interchange structure is adopted; Plan the take-off and landing paths through the take-off and landing point path model, using a conical space structure; Use the terminal route model to complete accurate navigation in the mission area, connecting the trunk route network with the specific mission location; According to the hierarchical structure of the multi-level route network, select the appropriate route level for planning.

7. The method for constructing a multi-level low-altitude air route network in a complex urban environment as claimed in claim 6, characterized in that: Step S72 includes: S721 sorts the high-priority aircraft in descending order of priority; S722 plans a path for each high-priority aircraft in order of priority, including: Use a global path planning algorithm to find the fastest path from the start point to the end point in the airspace three-dimensional route model and multi-level route network, while considering the flight time and risk assessment function R; Considering the trajectory and time schedule of the planned high-priority aircraft, evaluating the potential conflict between the path of the current high-priority aircraft and the planned high-priority aircraft; If there is a risk of conflict, adjust the flight plan of the current high-priority aircraft, including time adjustment, altitude adjustment, and path replanning; Under the premise of satisfying all constraints, select the optimal path with the shortest expected arrival time; S723 records the planned path and schedule of the current high-priority aircraft, and updates the trajectory information of the high-priority aircraft; S724 repeats steps S722-723 until the path planning of all high priority aircraft is completed.

8. The method for constructing a multi-level low-altitude air route network in a complex urban environment as claimed in claim 7, characterized in that: The path optimization objective function of the first optimization algorithm is as follows: R(x,y,z,t)=ω1·O(x,y,z)+ω2·P(x,y,z,t)+ω3·W(x,y,z,t)+ω4 T(x, y, z, t) Where J1 is the total path cost, E1(t) is the energy consumption function of the high priority aircraft, λ is the risk weight factor, t0 and t f are the start and end time, (x(t), y(t), z(t)) is the spatial position of the aircraft at time t, R(x, y, z, t) represents the comprehensive risk value at the spatial position (x, y, z) and time t, O(x, y, z) is the obstacle risk function, P(x, y, z, t) is the population density risk function, W(x, y, z, t) is the meteorological risk function, T(x, y, z, t) is the communication risk function, ω1, ω2, ω3, ω4 are dynamically adjusted weight coefficients.

9. The method for constructing a multi-level low-altitude air route network in a complex urban environment as claimed in claim 7, characterized in that: Step S73 includes: S731 builds a multi-objective path planning model based on the mission requirements, starting points and destinations of all medium and low priority aircraft, with optimization objectives including flight distance, energy consumption and route congestion; S732 considers the constraints of the three-dimensional airspace route model, as well as the paths and time scheduling of planned high-priority aircraft, and establishes collision avoidance constraints and airspace accessibility constraints; S733 performs iterative path optimization for medium and low priority aircraft, including: The flight path is represented as a chromosome code based on the route node sequence; the initial population is generated based on the shortest path algorithm; A multi-objective optimization function is used as the fitness function, taking into account the flight distance, energy consumption and route congestion; An elite retention strategy is used for selection, and a partial mapping crossover method is used, combined with adaptive mutation probability for optimization; After the iteration reaches the preset conditions, S734 outputs the optimal path sequence and completes the path planning of medium and low priority aircraft.

10. The method for constructing a multi-level low-altitude air route network in a complex urban environment as claimed in claim 9, characterized in that: The multi-objective optimization function in the second optimization algorithm is as follows: J2=w1·D+w2·E2+w3·C Where J2 is the comprehensive cost function, D is the flight distance, dist(n i , n i+1 ) represents node n i and n i+1 The spatial distance between them, E2 is the energy consumption, e(n i , n i+1 ) is the aircraft from node n i to n i+1 The energy consumed, C is the route congestion, γ(n i ) is node n i The route congestion evaluation function at is, N is the number of nodes, w1, w2, w3 are weight coefficients; Route congestion evaluation function γ(n i ) is calculated as follows: In the formula, Q(n i ) is node n i The number of aircraft passing per unit time, Q max For node n i The maximum throughput capacity, S(n i ) is node n i The average size of the aircraft at max is the maximum size of the aircraft, U(n i ) is node n i The cumulative priority penalty at max is the maximum value of the priority penalty.

Images