CN109814598B - Unmanned aerial vehicle low-altitude public navigation network design method - Google Patents

Abstract

The invention discloses a method for designing a low-altitude public route network of an unmanned aerial vehicle, which belongs to the technical field of route planning. The method includes acquiring the low-altitude flight environment data of the UAV in the planning area; obtaining the UAV airport layout information according to the UAV low-altitude flight environment data, and the UAV airport layout information includes: selection of a plurality of UAV airports. site and the service scope of each UAV airport; according to the UAV airport layout information, the UAV low-altitude flight environment data and the ant colony algorithm to obtain a plurality of three-dimensional routes; Form a low-altitude public air route network for drones. The present invention realizes the construction of the low-altitude public air route network of the unmanned aerial vehicle through the above technical scheme, and the route search efficiency is higher and the time consumption is shorter when the route is constructed.

Description

Design method of UAV low-altitude public route network

technical field

The invention belongs to the technical field of unmanned aerial vehicles, and particularly relates to a design method for a low-altitude public airway network of an unmanned aerial vehicle.

Background technique

The trajectory planning of UAV depends on the space division method and the trajectory planning algorithm. At present, the division of track planning space mainly adopts unit decomposition, which refers to an efficient method to discretize the airspace environment of UAV flight. This method divides the environment into free units and obstacle units, and uses a discrete method to represent the environment; unit; The representation methods include Voronoi diagram method and grid method. The Voronoi diagram builds an initial set of optional paths or sets up navigation nodes, and then selects the appropriate path through an optimization algorithm. The disadvantage is that the determination of the location and number of navigation nodes often requires repeated deliberation. The construction accuracy of the Voronoi diagram determines the optimization of the track cost. The accuracy is poor, and the adaptability to emergencies is poor; the grid-based spatial division can effectively solve this problem and has a wide range of applications. After building the trajectory planning space, use the path search algorithm to find the optimal trajectory. Among them, the heuristic algorithm is the most widely used, mainly including artificial potential field method, A* algorithm, fast search random tree algorithm, Dijkstra algorithm and in recent years. The emerging intelligent bionic algorithms such as ant colony algorithm, particle swarm algorithm and genetic algorithm. Ant colony algorithm successfully and efficiently solves the traveling salesman problem, and has advantages in solving complex optimization problems, especially discrete optimization problems, but it also has problems such as slow convergence speed and easy to fall into local optimum.

UAV trajectory planning is different from UAV low-altitude route planning. In terms of airspace effectiveness, trajectory planning is mostly one-time use, and its airspace effectiveness is also terminated after the task is completed, while the airspace involved in public routes is It remains unchanged for a long time, which can effectively improve and standardize the utilization of low-altitude airspace resources and facilitate UAV traffic safety management and control; from the perspective of planned space objects, track planning is oriented to lines, while public route planning is oriented to space In terms of the elements of the planning space environment, the trajectory planning only considers low altitude, terrain, and electromagnetic interference or missile danger areas, while the public route planning also considers the airspace on this basis. Policies, cellular networks, and densely populated areas are more closely related to human activities; from the perspective of service objects, track planning is more purposeful, mostly task-oriented, and is generally used for surveying, mapping, line patrol and other purposes, while public routes The planning needs to consider multiple application purposes, and the versatility is strong. Therefore, based on the characteristics of the above UAV low-altitude public routes, a design method of the UAV low-altitude public route network is proposed.

SUMMARY OF THE INVENTION

In order to solve the above problems, the present invention provides a method for designing a low-altitude public airway network for unmanned aerial vehicles, which includes: acquiring low-altitude flying environment data of unmanned aerial vehicles in a planning area, and the low-altitude flying environment data of unmanned aerial vehicles includes: terrain data, Low-altitude climate data, airspace policy data, and low-altitude mobile communication signal spatial distribution data; according to the low-altitude flight environment data of the UAV, the UAV airport layout information is obtained, and the UAV airport layout information includes: multiple UAV airports The site selection site and the service scope of each UAV airport; according to the UAV airport layout information, the UAV low-altitude flight environment data and the ant colony algorithm to obtain a plurality of three-dimensional routes; The three-dimensional route forms a low-altitude public route network for UAVs.

In the above method, preferably, obtaining a plurality of three-dimensional routes according to the UAV airport layout information, the UAV low-altitude flight environment data and the ant colony algorithm specifically includes: The service scope of the man-machine airport and the preset route level determine the two drone airports in each three-dimensional route included in each route level; according to the two drone airports of a three-dimensional route and the corresponding The low-altitude flight environment data of the UAV in the planning area is used to construct a mathematical model of the low-altitude flight environment of the UAV; the digitized mathematical model of the low-altitude flight environment of the UAV is sliced horizontally at a preset flight height to obtain the preset flight. A digital two-dimensional low-altitude environment mathematical model in height; based on the ant colony algorithm, a two-dimensional path search is performed in the digital two-dimensional low-altitude environment mathematical model to obtain a two-dimensional route, wherein, in the ant colony algorithm, search The space is formed by buffering a variable distance with the connection between the starting node and the ending node, and the buffer distance gradually increases from the initial value with 1 search step until there is an optimal route solution in the search space; according to the two-dimensional Route and reference terrain data to obtain a three-dimensional route; traverse two of the UAV airports in each three-dimensional route of all route levels to obtain multiple three-dimensional routes.

In the above-mentioned method, preferably, forming a low-altitude public air route network of the UAV according to a plurality of the three-dimensional routes specifically includes: according to the division heights of different route levels, the routes of the same level are used when the UAV shakes hands. UAVs with different altitudes, different priority levels of the same-level routes, and multiple three-dimensional routes form a low-altitude public route network of UAVs.

In the above-mentioned method, preferably, obtaining the UAV airport layout information according to the UAV low-altitude flight environment data specifically includes: using a maximum coverage location model according to the UAV low-altitude flight environment data , obtain the initial UAV airport layout information, the initial UAV airport layout information includes: the initial site selection sites of multiple UAV airports and the initial service scope of each UAV airport; Whether there is a cross-conflict of air routes at the address site, if it is judged to be yes, the initial UAV airport layout information is optimized to obtain the UAV airport layout information, otherwise the initial UAV airport layout information is used as UAV airport layout information.

In the above method, preferably, in the ant colony algorithm, the heuristic function η _ij (t):

分别为归一化后的下一节点离起始节点和终止节点的距离，d_Oi为起始节点和当前节点i间距离，d_OE为起始节点和终止节点间距离，C和ρ为常数，

表示下一节点与起始节点间距离的权重，ρ表示开始引入方向信息的路径长度阈值，allowed_k表示蚂蚁可以到达的节点。Among them, d _ij represents the distance between the current node i and the next node j,

are the distances between the normalized next node and the starting node and the ending node, respectively, d _Oi is the distance between the starting node and the current node i, d _OE is the distance between the starting node and the ending node, and C and ρ are constants ,

represents the weight of the distance between the next node and the starting node, ρ represents the path length threshold for starting to introduce direction information, and allowed _k represents the node that the ants can reach.

In the above method, preferably, the method further includes: in the ant colony algorithm, if the random number rand<max(P _i ) generated by the random function, use the roulette algorithm to determine the next Nodes are selected; if the random number rand=max(P _i ) generated by the random function, the node with the largest transition probability is taken as the next node.

In the above method, preferably, the obtaining of the three-dimensional route according to the two-dimensional route and the reference terrain data specifically includes: judging whether the terrain height of the next waypoint in the reference terrain data is greater than the height of the lower interface of the route and the minimum The difference of the obstacle clearance margin. If the judgment is yes, the height of the next waypoint is equal to the sum of the average grid height in the route range and the minimum obstacle clearance margin. If the judgment is no, the height of the next waypoint is equal to the The height of the interface under the above-mentioned route.

In the above method, preferably, after the determination is no, the method further includes: determining whether the next waypoint is in the mountain range in the reference terrain data and whether the height of the current waypoint is higher than the highest point in the mountain range. point and whether the height of the next waypoint is less than the height of the current waypoint, if it is judged to be yes, the height of the next waypoint is equal to the height of the current waypoint; if it is judged to be no, then jump to the next step A current waypoint altitude is equal to the lower interface altitude of the route.

In the above-mentioned method, preferably, during the two-dimensional path search, the first resolution is used, and the first resolution is lower than the second resolution, and the second resolution is adopted by the terrain data of mountainous areas. After obtaining the two-dimensional route and before obtaining the three-dimensional route according to the two-dimensional route and the reference terrain data, the method further includes: judging whether two adjacent waypoints of the two-dimensional route correspond to mountainous areas , and if it is determined to be yes, add several waypoints between the two adjacent waypoints.

In the above method, preferably, the UAV low-altitude flight environment mathematical model is constructed according to the two UAV airports of a three-dimensional route and the UAV low-altitude flight environment data in the corresponding planning area, specifically Including: constructing an initial model of the low-altitude flight space of the UAV, in the initial model of the low-altitude flight space of the UAV, the lower interface of the flight space is determined by the terrain data corresponding to the two UAV airports, and the upper interface of the flight space is determined by The spatial distribution of low-altitude mobile communication signals is determined; the elements that restrict the safe flight of the drone in the initial model of the low-altitude flight space of the drone are mathematically modeled, and the mathematical model of the low-altitude flight environment of the drone is completed. The elements of human-machine safe flight include: mountain constraints, high-rise building constraints, low-altitude climate constraints, and airspace policy restricted area constraints; correspondingly, in the ant colony algorithm, the constraints are determined in the local search space using the constraints Whether the proportion of obstacles represented by the elements of UAV safe flight and the communication blind area constraint elements is lower than the preset proportion threshold, if it is judged to be lower than the preset proportion threshold, adjust the search step size.

The beneficial effects brought by the technical solutions provided in the embodiments of the present invention are:

This paper proposes how to construct a low-altitude public route network for UAVs, and the route search is more efficient and time-consuming when building routes.

Description of drawings

1 is a schematic flowchart of a method for designing a low-altitude public airway network for an unmanned aerial vehicle according to an embodiment of the present invention;

2 is a schematic diagram of a search space provided by an embodiment of the present invention;

3 is a schematic flowchart of a method for obtaining a two-dimensional route according to an embodiment of the present invention;

4 is a schematic flowchart of a method for obtaining a three-dimensional route according to an embodiment of the present invention;

5 is a schematic structural diagram of a route model provided by an embodiment of the present invention;

6 is a schematic diagram of a route interval provided by an embodiment of the present invention;

7 is a schematic diagram of a flight interval provided by an embodiment of the present invention;

8 is a schematic diagram of a transition between routes at different levels according to an embodiment of the present invention;

FIG. 9 is a schematic diagram of the operation of the UAV airport, the approach route and the departure route (A) in FIG. 8 .

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

Referring to FIG. 1 , an implementation case of the present invention provides a method for designing a low-altitude public airway network for an unmanned aerial vehicle, which includes the following steps:

Step 101: Obtain low-altitude flight environment data of the UAV in the planning area.

Among them, the low-altitude flight environment data of UAVs (or the basic database of UAV route planning) consists of factors that affect the safe flight of UAVs, and UAV flight is closely related to the ground environment and human activities. Therefore, the influencing factors mainly include Basic geography, low-altitude climate, low-altitude communication environment, airspace policy, etc. Basic geographic data mainly includes topographic data, water system distribution, ground road network distribution and population distribution data. Low-altitude climate data: UAVs are active in the troposphere, and low-altitude climate has an important impact on their take-off, landing, operation and flight. Among them, the major weather phenomena are wind shear, thunderstorms, ice accretion, and low visibility caused by fog, haze, and sandstorms. Airspace policy data: It mainly includes the no-fly zone, restricted zone and danger zone for drones stipulated by the government, especially the protection scope of the obstacle restriction surface of civil aviation airports. The low-altitude mobile communication signal spatial distribution data (or low-altitude communication environment data) is represented by the mobile base station data: the UAV receives and sends radio signals to communicate with the ground station through the airborne link equipment, and receives the command and control of the ground control station. And task instructions, send its own attitude information and related task information. With the rapid development of the UAV industry, the UAV communication link presents a development trend that is closely integrated with cellular mobile communication technology, forming a "networked UAV". Therefore, ensuring that the drone operates within the coverage of the cellular network is critical to its safe flight. The implementation case of the present invention establishes a low-altitude flight cellular network environment for UAVs based on the distribution data of mobile base stations and their signal coverage. The planning area can be the whole of China, or it can be a certain area, such as North China, Central China, East China, etc.

Step 102 , obtaining the UAV airport layout information according to the low-altitude flight environment data of the UAV. The UAV airport layout information includes: site selection sites of multiple UAV airports and the service scope of each UAV airport.

Specifically, the drone airport refers to the organic whole composed of drone airports and related service facilities with legal airspace, and its hardware facilities include but are not limited to: drone runways, drone hangars, drones Assembly and debugging area, navigation and communication facilities, etc.; software facilities include but are not limited to: flight supervision system, air traffic control coordination notification system, etc. The UAV airport is the hub between the low-altitude route and the ground of the UAV. It can be used as the take-off and landing point and transfer site of the UAV to provide guarantee for the safe flight of the UAV. The location of the drone airport must have a wide field of view, good communication conditions, no high-rise buildings or mountains, and not within the restricted area of the airspace policy. According to the low-altitude flight environment data of the UAV, and based on the distribution of the existing ground transportation hubs, the UAV airport layout information is obtained by using the maximum coverage location model. site) and the initial service scope of the drone airport (preliminary service scope). During application, according to the service scope of air routes at all levels, fully consider the spatial distribution characteristics of factors that affect the safe flight of UAVs, such as population factors and terrain factors, based on the distribution of existing ground transportation hubs, and use the maximum coverage location model to select preliminary sites. As a UAV airport supporting the UAV airway traffic network.

The UAV airport layout obtained by using the maximum coverage location model does not consider the impact of cross-conflict on airway safety, and there are hidden dangers in flight safety, which cannot meet the actual flight requirements of UAVs. Therefore, the method also includes a location optimization step: optimizing the UAV airport layout. The optimized algorithms are: merging adjacent UAV airports, merging routes, collinear adjustment, low utilization route adjustment, no crossing, non-straight line coefficients, etc. That is to say, it is judged whether there is a cross conflict in the route connection between the UAV airports in the UAV airport layout. If it is judged to exist, the site selection optimization step, the field investigation and the low-altitude networking test step are performed in sequence, otherwise, the field investigation and low-altitude networking test steps are performed. Network test steps.

Field research and low-altitude networking test steps: After getting the preliminary UAV airport layout, it is necessary to conduct on-site inspection (or field research) to determine the specific location of the UAV airport. The following conditions must be met for site selection:

Airspace conditions: The risk of drones breaking into the air restricted area should be considered when building a drone airport in the restricted air area, and when building a drone airport in the area adjacent to the air restricted area.

Geographical conditions: The ground is open, free of obstacles, and there is enough space for the construction of the UAV runway; the geological unfavorable areas, possible submerged areas, active fault areas, mining areas, environmental and ecological protection areas, tourist attractions and cultural relics protection should be fully considered the influence of factors such as the region;

Communication conditions: meet the UAV safety flight communication link indicators;

Meteorological conditions: The impact of adverse meteorological conditions such as wind field, thunderstorm, ice accretion, and visibility on UAV flight safety should be fully considered;

Noise-sensitive area: It should be fully considered whether the aviation activity area meets the requirements of noise control indicators in the surrounding area;

Land use: should meet the requirements of relevant land use policies and regulations;

The electromagnetic environment is complex and dangerous: the impact of the space electromagnetic environment on the communication and navigation activities of the airport and the electromagnetic waves generated by the aviation activities on the sensitive facilities on the ground should be fully considered.

Considering the importance of the low-altitude communication environment to UAV control and safe flight, cellular network low-altitude coverage tests and business tests should be carried out in the area where the airport is located, such as electronic fence updates, real-time flight data reporting, and flight management command reception, etc. to ensure The UAV airport meets the UAV's safe flight communication link indicators (Civil Aviation Administration of China, "Low-altitude Networked UAV Safety Flight Test Report", 2018.1).

Step 103, obtaining a plurality of three-dimensional routes according to the UAV airport layout information, the UAV low-altitude flight environment data and the ant colony algorithm.

Specifically, this step includes: determining two drone airports in each three-dimensional route included in each route level according to the service range of each drone airport and a preset route level; The low-altitude flight environment data of the UAV in the man-machine airport and the corresponding planning area are used to construct a mathematical model of the low-altitude flight environment of the UAV. Set a digital two-dimensional low-altitude environment mathematical model at the flight height; perform a two-dimensional path search in the digital two-dimensional low-altitude environment mathematical model based on the ant colony algorithm, and obtain a two-dimensional route, in which, in the ant colony algorithm, the search space consists of The connection between the start node and the end node is used as a buffer with a variable distance, and the buffer distance gradually increases from the initial value with 1 search step until there is an optimal route solution in the search space; according to the two-dimensional route and reference terrain data, Get a 3D route; traverse two UAV airports in each 3D route at all route levels to get multiple 3D routes.

Among them, the UAV low-altitude route refers to the pre-planned air passage with a certain width for the UAV to fly below the minimum flight height of the manned aircraft. The actual route of the drone is called the route, and the route of the drone flying along the route is the centerline of the route. The purpose of delineating air routes is to maintain and standardize low-altitude traffic order, improve the utilization of low-altitude airspace resources, and ensure aviation and public safety. According to the positioning and service of the low-altitude route of the UAV, the present invention divides it into four levels: the backbone route, the main route, the branch route and the terminal route, that is, the preset route level is 4.

Among them, when the planning area is the whole of China, the low-altitude backbone route is the route connecting the capital with the capitals of provinces, autonomous regions and municipalities directly under the Central Government, and the routes connecting major economic centers, port and station hubs, commodity production bases and strategic locations; low-altitude backbone routes Refers to provincial low-altitude air routes with provincial political and economic significance; low-altitude branch routes refer to low-altitude routes connecting regions, mainly carrying the connection between unmanned aircraft airports and backbone/main routes; low-altitude terminal routes refer to connecting branch routes to Low-altitude routes between end users or one end user to another end user, which mainly carry the connection of unmanned aircraft from branch lines to terminal service points/stations such as logistics and catering delivery, or are deployed in complex terrain (such as mountainous areas, sparsely populated areas, etc.) low-altitude routes. The definitions of the low-altitude backbone route, the low-altitude backbone route, the low-altitude straight route and the low-altitude terminal route can also be adjusted according to the actual application and the specific scope of the planning area.

When applied, the routes can be classified and named by the combination of the first letter and the number, for example: the letters represent the routes at all levels, the backbone route is GG, the main route is ZG, the branch route is ZX, and the end route is MD; the third digit identifies the route. The route direction, such as 3 means north-south direction, 2 means east-west direction.

After the route levels are divided, the three-dimensional routes included in each route level are determined according to the service scope of each UAV airport, so as to determine the two UAV airports in each three-dimensional route. As the starting point and ending point of the path, and the low-altitude flight environment data of the UAV in the planning area corresponding to the two UAV airports, the UAV flight environment book sequence model is as follows:

First, construct the initial model of the low-altitude flight space of the UAV: Assuming that the coordinates of the i-th waypoint are (x _i , y _i , z _i ), the mathematical expression of the initial model of the low-altitude flight space can be:

where x _i is the longitude, y _i is the latitude, _zi is the height, and x _min , x _max , y _min and y _max represent the planning space plane range, which covers the site selection sites of the two drone airports , f _1i (x _i , y _i ) is the reference terrain altitude at the position of the ith waypoint, and f _2i ( _xi , y _i ) is the maximum communication height at the position of the ith waypoint.

In this initial low-altitude flight space model, the factors that determine the range of the UAV's flight height include: reference terrain and the spatial distribution of mobile communication base station signals. Specifically, the lower interface of the flight space is determined by the reference terrain, which reflects the terrain fluctuation of the route planning space and directly affects the minimum safe height of the route planning space. The upper interface of the flight space is determined by the signal space distribution of the mobile communication base station.

The mathematical expression of the base terrain is as follows:

Among them, x and y are the coordinates of the point where the reference terrain model is projected on the horizontal plane, and z is the elevation value corresponding to the point on the horizontal plane. a, b, c, d, e, g are constant coefficients used to control the datum terrain fluctuations in digital maps. By determining different constant coefficients, different reference landform features are simulated as the reference terrain of the UAV flight environment.

Therefore, the expression of the base terrain height at the position of the i-th waypoint is:

The signal coverage (or spatial distribution) of a mobile communication base station is related to the signal propagation law, the occlusion of ground objects and the environmental electromagnetic radiation. The signal strength shows irregular changes with the horizontal and vertical distance from the antenna. After actual testing, the current mobile cellular network can meet the application requirements of the UAV industry in most scenarios below 120 meters, as well as the requirements for UAV safe flight business link indicators in most areas below 300 meters. With the development of mobile communication technology, such as the commercial expansion of 5G technology, the height can be extended to below 1000m. Therefore, research can realize real-time monitoring of UAV's route through mobile cellular networking. Based on the national cellular mobile communication network, a low-altitude communication environment is constructed, and route planning is carried out on this basis to ensure real-time, highly reliable, and low-cost UAV communication links within the route. According to "Electromagnetic Radiation Monitoring Instruments and Methods of Radiation Environmental Protection Management Guidelines" (HJ-T10.2-1996), the formula for calculating the radiation power density of a single base station antenna is as follows:

According to the relationship between power density and electric field strength:

Assuming that the position of the j-th base station is (X _j , Y _j ), the electric field strength expression of the j-th base station at the position of the i-th waypoint can be obtained:

其中，P是基站天线发射功率(W)，G是基站天线增益(倍数)，θ是垂直面上与基站天线轴向的夹角，

为水平面上与基站天线轴向的夹角，

θ₁是天线下倾角，H是第j个基站天线离地高度(m)，以上参数均为常数；z_i是第i个航点离地高度(m)，R是第i个航点离第j个基站天线的水平距离(m)。

Among them, P is the base station antenna transmit power (W), G is the base station antenna gain (multiple), θ is the angle between the vertical plane and the base station antenna axis,

is the angle between the horizontal plane and the axis of the base station antenna,

θ ₁ is the downtilt angle of the antenna, H is the height of the jth base station antenna from the ground (m), and the above parameters are all constants; _zi is the height of the ith waypoint from the ground (m), and R is the distance of the ith waypoint from the ground. The horizontal distance (m) of the jth base station antenna.

Assuming that there are n base stations at the location of the i-th waypoint that can provide communication services, the field strength at the location of the i-th waypoint is:

The maximum communication altitude at the location of the i-th waypoint is: when E _i =C, the value of _zi , where C is the minimum value of the base station field strength that satisfies the safe flight of the UAV.

Secondly, mathematically model the elements that restrict the safe flight of the UAV in the initial model of the low-altitude flight space of the UAV, and complete the construction of the mathematical model of the low-altitude flight environment of the UAV.

Specifically, after constructing the initial model of the low-altitude flight space of the UAV, mathematical modeling is performed on the elements in the flight space that restrict the safe flight of the UAV, so as to build a mathematical model of the low-altitude flight environment of the UAV. Constraints mainly include: mountain constraints, high-rise building constraints, low-altitude climate constraints and airspace policy constraints. The mathematical model of each constraint element is as follows:

(1) Mathematical model of mountain constraining elements: The flying height of the UAV is low, and the natural elements on the surface, especially the mountain, are very important for its safe flight. The following formula can be used to simulate the prominent mountain and build a terrain environment model for UAV path search :

In the formula, z ₁ (x, y) is the elevation function of the peak, (x _j , y _j ) is the center coordinate of the j-th peak, H _j is the reference terrain height, a and b are the j-th peak along the x-axis, respectively. and the amount of attenuation in the y-axis direction to control the slope.

(2) Mathematical model of constraint elements of high-rise buildings: a cylinder can be used to represent a high-rise building, and its mathematical model can be expressed as:

Among them, z ₂ (x, y) is the elevation function of the high-rise building, (x _j , y _j ) is the center coordinate of the j-th building, H _j is the height of the j-th high-rise building, and R is the occupation of the high-rise building. ground radius.

(3) Mathematical model of low-altitude climate constraints: UAVs operate in the troposphere, and low-altitude climate has an important impact on its take-off, landing, operation and flight. Ice, as well as fog, haze, sandstorms and other weather phenomena that can cause low visibility. When planning the UAV route, the distribution of the above-mentioned areas with frequent occurrence of severe weather, such as the distribution of high-incidence areas of various climatic phenomena in the past two decades, is used as a reference. The research uses the near cylinder to judge the influence scope of the atmospheric environment, and its mathematical model can be expressed as:

d _ij =(xx _j ) ² +(yx _j ) ²

Among them, z ₃ (x, y) is the height function of the low-altitude climate-affected area, H is the upper limit of the height of the route planning area, (x _j , y _j ) is the center coordinate of the jth low-altitude climate-affected area, and d _ij is the ith The distance from the waypoint to the jth climate-affected area, d _min is the core area of the climate-affected area, the damage probability of the UAV in this area is 1, and d _max is the maximum radius affected by the atmosphere.

(4) Mathematical model of airspace policy restricted area constraints: Airspace policy restricted area constraints mainly refer to the no-fly areas, restricted-flying areas and dangerous areas of drones issued by local government departments. By introducing low-altitude airspace policies, the use of densely populated Areas and other airspace policy restrictions to build a safe and orderly low-altitude flight environment. The airspace policy restricted area can be derived from dynamic geo-fence data, and the airport plane range can be determined by the Civil Aviation Airport Obstacle Restriction Surface Protection Scope Data published by civil aviation. The lower limit is the ground surface, and there is no upper limit. Other restricted areas can be represented by the hemisphere model:

Among them, z ₄ (x, y) is the influence surface of the j-th restricted area, (x _j , y _j ) is the center coordinate of the j-th restricted area, and R is the footprint radius of the restricted area.

After the mathematical model of the low-altitude flight environment of the UAV is constructed, it is digitized, and then the digitized mathematical model of the low-altitude flight environment of the UAV is sliced horizontally at the preset flight height at the preset height to obtain the preset flight. The digital two-dimensional low-altitude environment mathematical model on the height, the steps are as follows:

First, digitize the mathematical model of the low-altitude flight environment of the UAV. The digitization method can use the grid method. The grid is in the form of a cube, and the length of the cube is consistent with the width of the route. Then obtain the preset flight altitude. In practical application, digitization can be realized based on technologies such as GIS mapping. The preset flight altitude can be obtained by the following methods: Fitting to obtain the initial altitude of the route according to the reference terrain data, and then calculating the sum of the initial altitude of the route and the minimum obstacle clearance margin of the UAV, so as to obtain the preset flight altitude. Then the digital UAV low-altitude flight environment mathematical model is sliced horizontally to obtain the two-dimensional low-altitude environment mathematical model of the UAV at the preset flight altitude.

After obtaining the digital two-dimensional low-altitude environment mathematical model at the preset flight height, a two-dimensional path search is performed in the digital two-dimensional low-altitude environment mathematical model based on the ant colony algorithm to obtain a two-dimensional route. The steps are as follows:

Based on the ant colony algorithm, in the digital two-dimensional low-altitude environment mathematical model, two UAV airports are used as the starting node and the end node to search for two-dimensional paths, and the two-dimensional routes corresponding to the two UAV airports are obtained.

In the traditional ant colony algorithm, with the expansion of the search space, the problem of "combination explosion" is prone to occur, which leads to the reduction of search efficiency. Therefore, in this step, the search space of the ant colony algorithm is formed by buffering the variable distance with the connection between the start node and the end node, so that the size of the search space can be reasonably controlled, which not only satisfies the optimal path solution in the search space In addition, the search efficiency is improved as much as possible, and the formed search space is shown in Figure 2. The buffer distance gradually increases with 1 search step from the initial value until there is an optimal route solution in the search space.

In order to improve the efficiency of path search, variable search step size is used in the search process, that is, the search step size is determined according to the proportion of obstacles in the local search space. When the proportion of obstacles in the search space is lower than the preset proportion threshold, the search step size is adjusted. , if the search step is longer than the previous search step, for example: when the proportion of obstacles in the search space is higher than 20%, the search step is 1; if it is less than 5%, the search step is 2. Obstacles can be characterized by the elements that restrict the safe flight of UAVs and the constraints of communication blind spots, and can also be characterized by the elements that restrict the safe flight of UAVs, the constraints of communication blind spots, and the airport elements. The elements that restrict the safe flight of UAVs can include airspace policy restricted area elements, mountain peak constraints, high-rise building constraints, and low-altitude climate constraints. The mountain constraints and high-rise building constraints can be collectively referred to as terrain features. That is to say, the factors that quantify the factors that restrict the safe flight of UAVs affect the potential risks of UAV flying, which is convenient for subsequent calculation of the cost-cost weighted attribute value of each spatial grid. The result is that when the UAV passes through each grid the potential risk, so that the potential risk factor is taken into account when obtaining the two-dimensional route. In order to simplify the calculation during quantification, the constraints of communication blind spots, high-rise building constraints, airspace policy restrictions, and mountain constraints are all transformed into "obstacles" in the flight environment, and the grid cost is 1. In the path search The principle of "no fly-in" is implemented; the atmospheric environment (ie, low-altitude climate constraints) is frequently reported data, so the "warning to fly-in" principle is followed in the path search, and the grid cost value ranges from 0.2 to 0.8, which is close to climate events. The closer the center is, the greater the cost; other areas are optional areas in the path search, and the grid cost is 0.

In the ant colony algorithm, the transition from the current node to the next node needs to calculate the transition probability of the current node to all adjacent nodes. The calculation formula is the prior art and is as follows:

为蚂蚁从当前节点i点到下一节点j点的转移概率，allowed_k表示蚂蚁可以到达的节点；α为信息素启发式因子，表示路径积累的信息素对路径选择的重要性程度，τ_ij(t)表示航迹段ij的信息素浓度；β为期望启发因子，表示启发因素对路径选择的重要性程度，η_ij(t)为启发函数，在现有技术中表示节点i、j之间距离的倒数，此种方法易导致蚂蚁贪图当前最短路径而陷入局部最优。在本发明实施例中将下一节点与起始节点和终止节点间的联系引入现有技术中的启发函数，以得到改进后的启发函数，如此能有效解决局部最优问题，同时提高算法效率。改进后的启发函数计算公式如下：in,

is the transition probability of the ant from the current node i to the next node j, allowed _k represents the node that the ant can reach; α is the pheromone heuristic factor, which represents the importance of the accumulated pheromone on the path selection, τ _ij (t) represents the pheromone concentration of the track segment ij; β is the expected heuristic factor, which represents the importance of the heuristic factor to path selection, η _ij (t) is the heuristic function, which represents the difference between nodes i and j in the prior art. The reciprocal of the distance between them is easy to cause the ants to covet the current shortest path and fall into the local optimum. In the embodiment of the present invention, the relationship between the next node and the starting node and the ending node is introduced into the heuristic function in the prior art, so as to obtain an improved heuristic function, which can effectively solve the local optimal problem and improve the algorithm efficiency at the same time. . The improved heuristic function calculation formula is as follows:

分别为归一化后的下一节点离起始节点和终止节点的距离，d_Oi为起始节点和当前节点i间距离，d_OE为起始节点和终止节点间距离，C和ρ为常数，

表示下一节点与起始节点间距离的权重，ρ表示开始引入方向信息的路径长度阈值，是为了避免蚂蚁在移动过程中过早受到方向信息影响而陷入局部最优，二者均在实际应用时确定数值。allowed_k表示蚂蚁可以到达的节点。Among them, d _ij represents the distance between the current node i and the next node j,

are the distances between the normalized next node and the starting node and the ending node, respectively, d _Oi is the distance between the starting node and the current node i, d _OE is the distance between the starting node and the ending node, and C and ρ are constants ,

Represents the weight of the distance between the next node and the starting node, and ρ represents the path length threshold at which direction information starts to be introduced, in order to avoid the ants being affected by the direction information prematurely during the movement process and falling into local optimum. Both are used in practical applications time to determine the value. allowed _k represents the nodes that the ants can reach.

In addition, in order to avoid the premature convergence problem of the ant colony algorithm, the embodiment of the present invention adopts the following algorithm to select the next node. The algorithm (or random roulette algorithm) is improved based on the roulette algorithm (or traditional roulette algorithm) and the greedy algorithm, which can not only improve the convergence speed of the traditional roulette algorithm, but also avoid the greedy algorithm easily falling into Defects of local optima. Specifically, if the random number rand<max(P _i ) generated by the random function (random function in matlab), the roulette method is used to select the next node, and this algorithm is equivalent to the roulette algorithm. (or traditional roulette algorithm), which can avoid the trap of local optimal value; if the random number rand=max(P _i ) generated by the random function, the node with the largest transition probability is selected as the next node. The algorithm is equivalent to the greedy algorithm and has a faster convergence rate. During application, by adding a random number (generated by matlab) that obeys a uniform distribution, rand∈[0, max(P _i )] is compared with P _i , and the selected probability P _i in the allowed _k set is greater than or equal to the local target of rand. The set of point i, the formula used is as follows, and finally the traditional roulette algorithm is used to filter out the optimal local target point.

Referring to Figure 3, the specific flow of this step is as follows:

Sub-step 1): Initialize the search space, determine the search step size according to the proportion of obstacles, and build an adjacency matrix.

Sub-step 2): Put the first generation of ants m (m=1, 2, ..., M) in the initial position, and add the initial position to the taboo table of each ant, where M represents the number of ants.

Sub-step 3): Find the nodes that can go to the next step, and form an optional node set LJD. Whether the distance between the current node and the termination node is less than the search step size or whether the set of optional nodes is an empty set; if not, execute sub-step 4), and if so, execute sub-step 6).

Sub-step 4): Calculate the state transition probability of the optional node. When calculating the heuristic function, the evaluation function of the A* algorithm is integrated, and the connection between the current node and the terminal node is introduced to optimize the search.

Sub-step 5): Select the next node (to_visit) using random roulette.

Sub-step 6): Update the node where the ant is located and the taboo table.

Substep 7): Record the path and its length.

Sub-step 8): repeat sub-step (2)-sub-step (7), until all ants of the first generation are traversed;

Sub-step 9): update pheromone;

Sub-step 10): repeat sub-step (2)-sub-step (9) until all algebras are traversed, K iterations;

Sub-step 11): get the optimal path. Here, it is necessary to calculate the path length and its cost attribute value (the following formula), and comprehensively consider the path length and the potential risk of UAV flying on this path to obtain the optimal path. The calculation process of the path cost cost attribute value is as follows:

w _i = max(w _{communication} , w _{restricted area} , w _airport , w _terrain , w _climate )

In the formula, W is the cost attribute value of the path, w _i is the cost attribute value of the ith node on the path, N is the number of nodes on the path, w _{communication} , w _{restricted area} , w _airport , w _terrain , w _climate respectively. Indicates various grid cost attribute values such as communication blind areas, restricted areas, airports, terrain and climate.

It should be noted that the term "node" in the embodiment of the present invention is the name during the path search process, and the "waypoint" is the name when the final path search result is obtained.

In practical application, it can be based on GeoSOT-3D earth grid subdivision technology, such as selecting the reference ellipsoid center in the CGC2000 geodetic coordinate system as the center of the subdivision ellipsoid to expand the longitude, latitude and height to -256°～256° respectively , -256°～256° and 0°～512°, form a space of 512°×512°×512°, and perform recursive octree division in three dimensions, forming a space down to the center of the earth and up to the surface Up to an altitude of 5000km, as large as the entire earth space, as small as a 0-32 level subdivision frame of centimeter-level blocks. At the same time, in order to efficiently organize and manage the spatial data, a unique code is given to the spatial position of any one-piece block at each level by using binary code according to the Z-sequence coding order, which is convenient for subsequent road searches.

On the basis of the two-dimensional route, the altitude information of each waypoint (that is, the waypoint elevation) is obtained, and then the three-dimensional waypoint coordinates are obtained, thereby obtaining the three-dimensional route. This step is shown in Figure 4, and the details are as follows:

The process of obtaining the waypoint elevation is as follows: determine whether the terrain height of the next waypoint in the reference terrain data is greater than the difference between the height of the interface under the route and the minimum obstacle clearance margin. The sum of the average grid elevation and the minimum obstacle clearance margin in the grid. If it is judged to be no, the height of the next waypoint is equal to the height of the lower interface of the route. In other words: the elevation of the waypoint is determined by the initial height of the lower interface of the route, the terrain height of the waypoint and the minimum obstacle clearance of the UAV. If the terrain height of the next waypoint is greater than the difference between the height of the lower interface of the route and the minimum obstacle clearance margin, the drone will climb, and the height of the next waypoint is equal to the sum of the terrain height and the minimum obstacle clearance margin; otherwise, the drone will climb The pitch angle remains unchanged, and the flight continues along the current altitude, that is, the altitude of the next waypoint is equal to the altitude of the current waypoint.

In order to safely cross the mountain, the UAV needs to climb and fly. When it crosses the highest point, it can continue to fly at the current altitude. Therefore, after the judgment is no, the method further includes: judging that the next waypoint is in the reference terrain data Whether it is in the mountain range and the current waypoint height is higher than the highest point of the mountain range and the next waypoint height is less than the current waypoint height, if it is judged as yes, the next waypoint height is equal to the current waypoint height; that is Say: When the highest point is crossed, the height of the next waypoint is compared with the height of the current waypoint. If the height of the next waypoint is lower than the height of the current waypoint, the height of the next waypoint is equal to the height of the current waypoint. This step belongs to the step of correcting the elevation of the waypoint. In other embodiments, the height of the next waypoint and the height of the current waypoint may also be compared after the height of the waypoint is acquired.

When the UAV passes through different terrain environments, its path search has different resolution requirements for the digital environment. For example, the terrain factors in the plain area have little effect, and only other natural or artificial features are considered, which requires lower resolution. When the drone passes through the mountainous area, the terrain factor has a great influence, and the resolution requirements are very high; however, as the resolution increases, the algorithm efficiency will decrease, so when searching for a two-dimensional path, the first resolution is used. The first resolution is lower than the second resolution, the second resolution is the resolution adopted by the terrain data of the mountainous area, and the first resolution is the resolution adopted by the terrain data of the plain area. After the two-dimensional route is obtained, and before the three-dimensional route is obtained according to the two-dimensional route and the reference terrain data, the method further includes: judging whether two adjacent waypoints of the two-dimensional route correspond to the mountainous area, and if it is judged to be yes, in the two-phase route Add several waypoints between adjacent waypoints. That is to say: first use a relatively low resolution (ie the first resolution) to perform a two-dimensional route search, and then encrypt the two-dimensional waypoint data based on high-precision terrain data (ie, the second resolution) for mountainous areas, That is to add several equally spaced waypoints between two adjacent waypoints to reduce the distance between adjacent waypoints. In other words, the two adjacent waypoints of the two-dimensional route are encrypted according to the terrain resolution. This step belongs to the step of encrypting the two-dimensional waypoints.

According to the aforementioned method for obtaining a three-dimensional route, two UAV airports in each three-dimensional route of all route levels are traversed to obtain a plurality of three-dimensional routes.

After obtaining multiple three-dimensional routes, in addition to the safety requirements, the UAV route also needs to meet the constraints of maneuverability parameters such as the UAV's dynamic turning rate, maximum climb angle, and maximum dive angle. Among them, dynamic turning requires that the curvature at any point on the route must be less than the maximum curvature that the UAV can achieve, and in three-dimensional space, the flightable path is determined by curvature and torsion. For UAV, the curvature of the path Equivalent to yaw rate turning, torsion rate is equivalent to roll rate roll. The maximum climb angle and the maximum dive angle are the maximum angles that the drone can continuously climb and dive in a single time in the vertical direction. If the angle is too large, the drone will stall. Therefore, in order to satisfy the dynamic constraints of the UAV, the three-dimensional route is optimized according to the UAV's maximum turning angle, maximum roll angle and maximum climb angle.

After the three-dimensional route is optimized, the method further includes: a simulated flight step and an actual flight test step.

After generating the route, first perform a system simulation flight on it to verify the flightability and safety of the route, which mainly includes the following steps:

(1) Real 3D data acquisition and modeling. The UAV is equipped with an ordinary optical camera to collect data on the route area, and through point cloud processing for 3D modeling, the real 3D environment of the survey area is obtained.

(2) UAV system modeling. The UAV system module includes functional modules such as the UAV body dynamics model, navigation, engine and control to achieve real-time solution of UAV flight parameters.

(3) Atmospheric environment modeling. Including atmospheric data simulation and atmospheric disturbance simulation. The atmospheric data simulation mainly includes atmospheric temperature, pressure and density, and the atmospheric disturbance simulation mainly includes the wind field.

After the simulated flight is qualified, the route generation area is flown on the spot, and the generated waypoints are imported into the UAV flight control, and the UAV flies according to the established route to verify the rationality and safety of the route.

It should be noted that the result obtained by the ant colony algorithm is an air route, and then a space body is formed with the air route as the center line. The shape of the space body may be a cylinder or other shapes, which are not limited in this embodiment. Due to factors such as ground navigation facilities, air traffic management, flight tasks, and terrain, a route is often composed of waypoints such as starting points, turning points, and ending points. Therefore, referring to Figure 5, a route consists of multiple line segments (flight segments). The expression and positioning of the route (flight segment) in the three-dimensional space are determined by factors such as the starting point O and the ending point E, the route angle a and the altitude H.

route = {start, destination, route angle, altitude}

route = {start, destination, route angle, altitude, route width}

The minimum safe altitude of the route is proposed to ensure the safe flight of unmanned aircraft on the route, and is determined by the relevant regulations of unmanned aircraft control, the performance constraints of unmanned aircraft, mission constraints and flight environment. The safe flight altitude of the aircraft is equal to the sum of the maximum elevation and the minimum obstacle clearance margin within the route range. The maximum elevation refers to the highest elevation of the ground object, and the minimum obstacle clearance margin refers to the minimum vertical separation that should be guaranteed when the aircraft is over obstacles. Influencing factors include meteorological conditions that may cause altitude deviations, instrument errors, and UAV performance.

Referring to Figures 6 to 7, the route interval D refers to the distance between two adjacent routes; the flight interval L is the distance of the aircraft based on time or space, which is divided into horizontal interval and vertical interval L _z , wherein the horizontal interval is further divided into side intervals. Toward spacing L _x and longitudinal spacing _Ly . If the flight safety of the aircraft is to be ensured, the route separation must be greater than the sum of the flight lateral safety separation and the route width W. Therefore, the route separation is determined by the flight safety separation of the unmanned aircraft. There are many factors that affect the flight interval, mainly including: 1) The influence of the drone itself. The communication, navigation, surveillance performance and its intervention capability (airborne collision avoidance capability and control capability) of the UAV system all play a key role in the flight safety separation; 2) the impact of natural environment. The natural environment has a very important impact on the flight safety of UAVs. For example, ice accretion will cause the aerodynamics of the wings to deteriorate, wind shear will affect the flight attitude of the UAV, and rain and snow will affect the visibility in the air, causing no noise. Man-machine system failure, etc.; 3) Route structure and traffic flow density. The increase in the complexity of the route structure and the density of traffic flow will lead to an increase in the risk of drone flight collisions, so the number of drones in a certain airspace at a certain point in time should be controlled. The route separation can be studied by risk collision modeling in the lateral, longitudinal and vertical directions of the route system based on the aircraft collision model theory.

Based on the existing low-altitude airspace opening policies and low-altitude airspace types, and comprehensively considering UAV communication requirements and route planning environmental constraints, this embodiment preliminarily defines the space height of route planning.

According to the "Interim Regulations on the Administration of Unmanned Aircraft Flight (Draft for Comment)", light UAVs can fly in the airspace below 120 meters without approval; the "Administrative Regulations on the Use of Low-altitude Airspace" defines low-altitude airspace below 1000m, which is divided into two categories: It is divided into three categories: control area (including visual flight route), reporting airspace and surveillance airspace. In order to ensure the flight safety of manned and unmanned aerial vehicles, low-altitude routes should be set up in the control area; the "Low-altitude Networked UAV Safety Flight Test Report" pointed out: Based on the national cellular mobile communication network (4G/5G technology), the current The mobile cellular network can meet the application requirements of the UAV industry in most scenarios below 120 meters, the requirements for UAV safety flight business link indicators in most areas below 300 meters, and the full coverage of communications below 1,000 meters.

Considering the above requirements comprehensively, take the true height of 120m and 1000m as the lower limit and upper limit of the UAV low-altitude high-speed route planning, respectively. The UAV backbone, trunk and branch routes are planned within this height range, and the end route is used as a low-speed route. Its altitude range is less than 120m, to the lowest safe altitude in the area, and it undertakes the transition function between the air route and the take-off and landing point, and solves the "last mile" problems such as community air routes.

Step 104, forming a low-altitude public route network of the UAV according to the multiple three-dimensional routes.

After obtaining the low-altitude public route network of UAVs, in order to build a safe and efficient airway traffic network, different planning altitude ranges are set for different levels of routes. Different types will lead to different minimum safe altitudes, so plan altitude ranges for each level of routes according to the minimum safe altitudes.

When the density of the route network is large enough, there will be intersection problems on the routes of the same level. Considering the flight characteristics of drones, the intersection of routes can easily lead to collisions of drones, which brings great risks to the safe flight and management of routes. Set different flight levels and route priorities for the same-level routes to solve the route intersection problem, as follows: 1) Priority principle: From the perspective of time, that is, UAVs with higher priority types pass the intersection first. , such as emergency rescue drones. 2) Layering of flight heights: Considering the space, that is, when the drone shakes hands, another altitude layer is temporarily called to avoid conflicts. That is, according to the different heights of different route levels, the same level of routes has different heights when the drones shake hands, the drones of the same level of routes have different priorities, and multiple three-dimensional routes form a low-altitude public route network for drones.

Referring to Figures 8 to 9, the transition of the UAV between different levels is realized through the UAV airport and its airspace, the arrival route, and the departure route. The specific process is as follows: starting point O → arrival route → branch/main route Airport ZG1 → branch/main route → backbone route Airport GG1 → backbone route → backbone route Airport GG2 → branch/main route → branch/main route Airport ZG2 → departure route → destination E.

To sum up, the beneficial effects of the embodiments of the present invention are as follows:

This paper proposes how to construct a low-altitude public route network for UAVs, and the route search is more efficient and time-consuming when building routes.

It is known from the technical common sense that the present invention can be realized by other embodiments without departing from its spirit or essential characteristics. Accordingly, the above-disclosed embodiments are, in all respects, illustrative and not exclusive. All changes within the scope of the present invention or within the scope equivalent to the present invention are encompassed by the present invention.

Claims (9)

1. A method for designing an unmanned aerial vehicle low-altitude public route network, characterized in that the method comprises: Acquire low-altitude flight environment data of the UAV in the planning area, and the low-altitude flight environment data of the UAV includes: terrain data, low-altitude climate data, airspace policy data and low-altitude mobile communication signal spatial distribution data; The UAV airport layout information is obtained according to the low-altitude flight environment data of the UAV, and the UAV airport layout information includes: the site selection sites of a plurality of UAV airports and the service scope of each UAV airport; According to the UAV airport layout information, the UAV low-altitude flight environment data and the ant colony algorithm to obtain a plurality of three-dimensional routes; Form a UAV low-altitude public route network according to a plurality of the three-dimensional routes; According to the UAV airport layout information, the UAV low-altitude flight environment data and the ant colony algorithm to obtain a plurality of three-dimensional routes, specifically including: Determine two of the drone airports in each three-dimensional route included in each route level according to the service range of each of the drone airports and the preset route level; Build a mathematical model of the UAV low-altitude flight environment according to the two UAV airports of a three-dimensional route and the data of the UAV low-altitude flight environment in the corresponding planning area; The digitized mathematical model of the low-altitude flight environment of the unmanned aerial vehicle is sliced horizontally at a preset flight height to obtain a digitized two-dimensional low-altitude environment mathematical model on the preset flight height; Based on the ant colony algorithm, a two-dimensional path search is performed in the digital two-dimensional low-altitude environment mathematical model to obtain a two-dimensional route, wherein, in the ant colony algorithm, the search space is connected by a line between the start node and the end node Buffer formation with variable distance is performed, and the buffer distance gradually increases from the initial value with a search step size until there is an optimal route solution in the search space; According to the two-dimensional route and the reference terrain data, a three-dimensional route is obtained; Traverse two of the UAV airports in each 3D route of all route levels to obtain multiple 3D routes; The process of constructing the mathematical model of the UAV low-altitude flight environment is as follows: First, construct the initial model of the low-altitude flight space of the UAV: Assuming that the coordinates of the i-th waypoint are (x _i , y _i , z _i ), the mathematics of the initial model of the low-altitude flight space is:

where x _i is the longitude, y _i is the latitude, _zi is the height, and x _min , x _max , y _min and y _max represent the planning space plane range, which covers the site selection sites of the two drone airports , f _1i (x _i , y _i ) is the reference terrain height at the position of the ith waypoint, f _2i ( _xi , y _i ) is the maximum communication height at the position of the ith waypoint; The reference terrain height at the position of the i-th waypoint is:

Secondly, mathematically model the elements that restrict the safe flight of the UAV in the initial model of the low-altitude flight space of the UAV, and complete the construction of the mathematical model of the low-altitude flight environment of the UAV; The elements that restrict the safe flight of the drone mainly include: mountain restraint elements, high-rise building restraint elements, low-altitude climate restraint elements and airspace policy restricted area restraint elements; The mathematical model of the mountain constraint elements is:

In the formula, z ₁ (x, y) is the elevation function of the mountain, x is the coordinate of the mountain in the x-axis direction, y is the coordinate of the mountain in the y-axis direction, and (x _j , y _j ) is the center coordinate of the jth peak , H _j is the reference terrain height, a and b are the attenuation of the j-th peak along the x-axis and y-axis, respectively, to control the slope; The mathematical model of the constraint elements of high-rise buildings is expressed as: z ₂ (x, y)=H _j ,

Among them, z ₂ (x, y) is the elevation function of the high-rise building, x is the coordinate of the high-rise building in the x-axis direction, y is the coordinate of the high-rise building in the y-axis direction, (x _j , y _j ) is the jth high-rise building The center coordinates of the building, H _j is the height of the j-th high-rise building, and R is the footprint radius of the high-rise building; The mathematical model of the low-altitude climate constraint elements is:

d _ij =(xx _j ) ² +(yx _j ) ² Among them, z ₃ (x, y) is the height function of the low-altitude climate-affected area, x is the coordinate of the low-altitude climate-affected area in the x-axis direction, y is the coordinate of the low-altitude climate-affected area in the y-axis direction, and H is the height of the route planning area Upper limit, (x _j , y _j ) the center coordinates of the jth low-altitude climate-affected area, d _ij is the distance from the i-th waypoint to the j-th climate-affected area, d _min is the core area of the climate-affected area, in The damage probability of the UAV in the core area is 1, and d _max is the maximum radius affected by the atmosphere; The mathematical model of the constraint elements of the airspace policy restricted area is:

Among them, z ₄ (x, y) is the influence surface of the j-th restricted area, (x _j , y _j ) is the center coordinate of the j-th restricted area, and R is the footprint radius of the restricted area. 2 . The method according to claim 1 , wherein the forming a low-altitude public air route network of UAVs according to a plurality of the three-dimensional routes specifically comprises: 2 . According to the different altitudes of different route levels, the same level of routes when the drones shake hands at different heights, the same level of routes with different priority levels of drones, and a plurality of the three-dimensional routes to form a low-altitude public route network for drones. 3. The method according to claim 1, wherein the obtaining of the UAV airport layout information according to the UAV low-altitude flight environment data specifically includes: According to the UAV low-altitude flight environment data, the maximum coverage site selection model is used to obtain initial UAV airport layout information, and the initial UAV airport layout information includes: initial location sites of multiple UAV airports and each the initial service scope of the drone airport; It is judged whether the initial site selection site has a cross conflict of routes, and if it is judged to be yes, then the initial unmanned aerial vehicle airport layout information is optimized to obtain the unmanned aerial vehicle airport layout information, otherwise, the initial unmanned aerial vehicle airport layout information is obtained. The aircraft airport layout information is used as the UAV airport layout information. 4. The method of claim 1, wherein In the ant colony algorithm, the heuristic function η _ij (t):

Among them, d _ij represents the distance between the current node i and the next node j,

are the distances between the normalized next node and the starting node and the ending node, respectively, d _Oi is the distance between the starting node and the current node i, d _OE is the distance between the starting node and the ending node, and C and ρ are constants ,

represents the weight of the distance between the next node and the starting node, ρ represents the path length threshold for starting to introduce direction information, and allowed _k represents the node that the ants can reach. 5. The method according to claim 4, wherein the method further comprises: In the ant colony algorithm, if the random number generated by the random function rand<max(P _i ), the roulette algorithm is used to select the next node; if the random number generated by the random function rand=max (P _i ), the node with the largest transition probability is taken as the next node. 6. The method according to claim 1, wherein the obtaining a three-dimensional route according to the two-dimensional route and the reference terrain data, specifically comprises: Determine whether the terrain height of the next waypoint in the reference terrain data is greater than the difference between the height of the lower interface of the route and the minimum obstacle clearance margin. The sum of the minimum obstacle clearance margins. If the judgment is negative, the altitude of the next waypoint is equal to the altitude of the lower interface of the route. 7. The method according to claim 6, wherein after the judgment is no, the method further comprises: Judging whether the next waypoint is in the mountain range in the reference terrain data and whether the height of the current waypoint is higher than the highest point of the mountain range and whether the height of the next waypoint is less than the height of the current waypoint, if it is judged to be yes, then the height of the next waypoint is equal to the height of the current waypoint; If the judgment is no, then jump to the next step where the current waypoint height is equal to the interface height under the route. 8 . The method according to claim 1 , wherein when searching for a two-dimensional path, a first resolution is used, the first resolution is lower than a second resolution, and the second resolution is a mountainous area. 9 . The resolution used for the terrain data; After the obtaining of the two-dimensional route and before the obtaining of the three-dimensional route according to the two-dimensional route and the reference terrain data, the method further includes: It is judged whether two adjacent waypoints of the two-dimensional route correspond to mountainous areas, and if it is judged to be yes, several waypoints are added between the two adjacent waypoints. 9 . The method according to claim 1 , wherein the drone low-altitude flight environment is constructed according to the two drone airports of a three-dimensional route and the drone low-altitude flight environment data in the corresponding planning area. 10 . Mathematical models, including: Build an initial model of the low-altitude flight space of the UAV. In the initial model of the low-altitude flight space of the UAV, the lower interface of the flight space is determined by the terrain data corresponding to the two UAV airports, and the upper interface of the flight space is moved by the low altitude. The spatial distribution of communication signals is determined; Carry out mathematical modeling on the elements that restrict the safe flight of the UAV in the initial model of the UAV low-altitude flight space, and complete the construction of the UAV low-altitude flight environment mathematical model, and the elements that restrict the safe flight of the UAV include: Mountain peak constraints, high-rise building constraints, low-altitude climate constraints, and airspace policy constraints; Correspondingly, in the ant colony algorithm, it is judged in the local search space whether the proportion of obstacles represented by the elements that constrain the safe flight of the UAV and the constraining elements of the communication blind area is lower than the preset proportion threshold, if it is judged as If it is lower than the preset ratio threshold, adjust the search step size.

Images