Low-altitude flight path planning method and system based on CIM

Abstract

The invention relates to the technical field of flight path planning, in particular to a CIM-based aircraft low-altitude flight path planning method and system. The method comprises the following steps: dividing a flight area based on a CIM platform to obtain a plurality of different airspace flight grids, and calculating and storing the surface height of the airspace flight grids; setting a flight path cost function, and selecting a path with the lowest cost as a planned path of the low-altitude aircraft; the method comprises the steps of firstly planning a rough flight path, then designing a detour path based on an obstacle, and then optimizing the detour path to obtain a planned flight path; a flight path is subjected to multi-wheel simplification processing, and unnecessary turning and bypassing are reduced; and overlapping the airspace control information and other temporary constraint conditions to obtain a final flight path. According to the method, the space grid and the navigation map of the flight area are generated based on the CIM platform, the low-altitude flight path is dynamically and rapidly planned, and the calculation efficiency of the algorithm is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of flight path planning, and particularly to a method and system for low-altitude flight path planning of an aircraft based on CIM. Background Art

[0002] The low-altitude area generally refers to the airspace with an altitude below 1000 meters. The main aircraft operating in this airspace are general aviation aircraft, such as unmanned aerial vehicles (UAVs); and the application altitude of civilian UAVs usually concentrates between 0 and 400 meters. The low-altitude area contains complex urban geographical environments, and there are many building obstacles when UAVs are flying. If a large number of aircraft fly frequently in the same airspace, collision accidents are likely to occur. Therefore, higher requirements are put forward for the flight path planning of aircraft.

[0003] Low-altitude flight path planning is the basis for UAVs to perform tasks, and the purpose is to plan the optimal path between the starting position and the target position in a specific environment. Common methods include graph search algorithms, linear programming methods of operations research, intelligent optimization algorithms, dynamic programming algorithms, algorithms based on graphics, heuristic algorithms, machine learning algorithms, etc. These existing common algorithms often face the following problems and challenges:

[0004] First, it is difficult to model complex environments. The low-altitude flight environment is complex and changeable, with various obstacles such as buildings, trees, and utility poles. Moreover, the shapes and distributions of these obstacles are irregular, and existing modeling methods are difficult to integrate rich data to accurately represent the characteristics of these complex environments.

[0005] Second, the algorithm efficiency and real-time performance are poor. In the face of path planning in a large-scale airspace, traditional path planning algorithms often adopt the strategy of traversing spatial points. Although a relatively optimal path can be found, the computational complexity of the algorithm is often high, requiring a large amount of computational time and resources. Especially when performing path planning for multiple aircraft in a large-scale low-altitude area, as the search space becomes larger, the computational time is further increased.

[0006] Third, it is easy to fall into local optimal solutions. Many intelligent algorithms adopt heuristic search strategies. Although the search efficiency is improved to a certain extent, this search strategy often makes decisions based on local information. For example, genetic algorithms search for optimal solutions based on simulating genetic variation and natural selection in the process of biological evolution; however, during the search process, due to the randomness of crossover and mutation operations, it may fall into a local optimal region and be unable to jump out. There are also some intelligent algorithms that perform flight path planning by rasterizing spatial regions, but their optimization ability is weak, and the planned paths have many turns and large turning angles.

[0007] Generally speaking, most of the existing algorithms are applicable to the management and control of drones in small-scale or typical scenarios, and cannot meet the requirements of large-scale path planning in low-altitude airspace. The data volume and types used in the existing methods are not perfect enough. The obstacles are mainly buildings and trees, and power poles, street lights, and other urban infrastructure are not considered, making the obstacle recognition and threat calculation less precise. Summary of the Invention

[0008] In view of the problems existing in the prior art, the present invention proposes a low-altitude flight path planning method and system based on the CIM platform, which generates a spatial grid and a navigation map of the flight area based on the CIM platform, realizes dynamic and rapid planning of low-altitude flight paths, and improves the calculation efficiency of the algorithm.

[0009] On the one hand, an embodiment of the present invention provides a CIM-based low-altitude flight path planning method, including the following steps:

[0010] S1. Divide the flight area based on the CIM platform to obtain multiple different airspace flight grids; calculate the surface height of each airspace flight grid and store the calculation results;

[0011] S2. Set a flight path cost function; select the path with the lowest cost as the planned path of the low-altitude aircraft among all possible flight paths;

[0012] S3. First plan a rough flight path, then design a detour path based on obstacles, and then optimize the detoured path to obtain the planned flight path;

[0013] S4. Perform multiple rounds of simplification processing on the flight path planned in step S3 to reduce unnecessary turns and detours and obtain an optimized flight path;

[0014] S5. Overlay airspace control information and other temporary constraint conditions to obtain the final flight path.

[0015] On the other hand, an embodiment of the present invention also provides a CIM-based low-altitude flight path planning system, which is implemented based on the above path planning method. The path planning system includes:

[0016] A data fusion and storage module for dividing the flight area based on the CIM platform to obtain multiple different airspace flight grids; calculating the surface height of each airspace flight grid and storing the calculation results;

[0017] A flight cost calculation function for setting a flight path cost function; selecting the path with the lowest cost as the planned path of the low-altitude aircraft among all possible flight paths;

[0018] A path planning module, which is used to first plan a rough flight path, then design a detour path based on obstacles, and then optimize the detour path to obtain the planned flight path;

[0019] A path optimization module, which is used to perform multiple rounds of simplification processing on the planned flight path to reduce unnecessary turns and detours and obtain the optimized flight path;

[0020] A condition overlay module, which is used to overlay airspace control information and other temporary constraint conditions to obtain the final flight path.

[0021] Compared with existing path planning algorithms, the present invention can give full play to the advantages of high data accuracy and high data capacity of the CIM platform based on the CIM platform. The CIM platform integrates various types of data within the flight area, such as terrain, buildings, traffic, Internet of Things data, etc., improving the perception accuracy of the environment, identifying various obstacles to the greatest extent, and realizing high-precision modeling of the flight environment. At the same time, in terms of the path specification algorithm, the present invention adopts a "coarse first and then fine" path planning idea, which can significantly improve the efficiency of path planning.

[0022] The advantages of the present invention are specifically embodied as follows:

[0023] 1. Considering the characteristic that the flight path requires quick direct access, the present invention proposes a fast planning method based on the initial straight-line path, and dynamically optimizes and adjusts local road sections through spatial grid features, thereby generating the final flight path. Compared with the prior art, this method can quickly generate an available flight trajectory, significantly reduce the amount of computation, and improve the calculation efficiency.

[0024] 2. The present invention quickly generates a spatial flight grid of the flight area based on the CIM platform, and at the same time establishes virtual space buffer entities for small objects such as utility poles and lamp posts; and uses the off-screen rendering technology of the CIM platform to obtain the surface elevation of the generated city, with advantages such as fast generation speed, convenient update, and high refinement degree.

[0025] 3. When dynamically and quickly planning the low-altitude flight path, the present invention first roughly plans the flight path, then bypasses obstacles and no-fly zones based on the flight navigation map, and finally uses a variety of algorithms to continuously optimize the flight path, thereby achieving the purpose of fast planning. And by defining the flight path cost function and considering factors such as distance, wind resistance, and safety, the best flight path planning is ensured.

[0026] 4. The present invention supports dynamically superimposing various control factors during the flight onto the flight navigation map. This mode can flexibly add diverse spatial management requirements, and various forms of airspace control conditions can be set, such as no-fly zones, attraction zones, safety buffer zones, etc., making the controllable details of the planned flight path richer and ensuring the safety of the flight path. Description of the Drawings

[0027] Figure 1 This is the overall flowchart of the CIM-based low-altitude flight path planning method in the embodiments of the present invention;

[0028] Figure 2 This is the depth map obtained by GPU off-screen rendering to obtain the depth buffer and perform visual rendering in the embodiments of the present invention;

[0029] Figure 3 This is the flight navigation map calculated during the flight path planning process in the embodiments of the present invention;

[0030] Figure 4 This is the flight navigation map that breaks the initial flight path into obstacle sections and straight flight sections in the embodiments of the present invention;

[0031] Figure 5 This is the schematic diagram of edge point recognition in the bypass algorithm in the embodiments of the present invention;

[0032] Figure 6 This is one of the schematic diagrams of the horizontal bypass route in the embodiments of the present invention;

[0033] Figure 7 This is the second schematic diagram of the horizontal bypass route in the embodiments of the present invention;

[0034] Figure 8 This is the schematic diagram of the flight path for climbing and bypassing obstacles in the embodiments of the present invention, where (a) is the flight path when the distance between obstacles is less than the threshold, and (b) is the flight path when the distance between obstacles is greater than the threshold;

[0035] Figure 9 This is the schematic diagram of directly connecting and simplifying the flight path based on any two key grid points in the embodiments of the present invention;

[0036] Figure 10 This is the schematic diagram of realizing flight path simplification based on line segment interpolation points in the embodiments of the present invention, where (a) is the schematic diagram of the flight path before simplification, (b) is the schematic diagram of the flight path after inserting internal interpolation points, (c) is the polyline simplification diagram based on internal interpolation points, and the solid line part of (d) is the effect diagram of the simplified flight path;

[0037] Figure 11 This is the schematic diagram of re-planning the wavy flight trajectory in the embodiments of the present invention, where (a) is the schematic diagram of the flight path before re-planning, and (b) is the schematic diagram of the flight path after re-planning;

[0038] Figure 12 This is the schematic diagram of the simplification process of the wave crest trajectory when re-planning the wavy flight trajectory, where (a) is the schematic diagram of the flight trajectory before simplification, and (b) is the schematic diagram of the flight trajectory after simplification;

[0039] Figure 13 Schematic diagram of storing airspace control area data into the airspace control grid table in the embodiment of the present invention;

[0040] Figure 14 Flight navigation map obtained after superimposing various airspace control conditions in the embodiment of the present invention. Detailed implementation manners

[0041] The following further describes the implementation manners of the present invention in detail in conjunction with embodiments and the accompanying drawings, but the implementation manners of the present invention are not limited thereto.

[0042] Embodiment

[0043] This embodiment provides a low-altitude flight path planning method based on CIM. In the process of low-altitude flight path planning, the overall technical process is as follows: First, calculate and store the airspace flight grids, then plan the flight path, then optimize the flight path, and finally superimpose the airspace control area and other constraint conditions to obtain the final flight path. As Figure 1 shown, specifically including:

[0044] S1. Divide the flight area based on the CIM platform to obtain multiple different airspace flight grids; calculate the surface height of each airspace flight grid and store the calculation results.

[0045] In this embodiment, this step includes the following steps:

[0046] S11. Integrate multi-source data of the urban real scene model based on the CIM platform.

[0047] Integrating multi-source data of the urban real scene model based on the CIM platform, including terrain, buildings, trees, and other equipment and facilities, and establishing virtual space buffer entities for small but potentially flight safety-affecting equipment and facilities such as wires and street lights.

[0048] According to the integrated data, establish a flight area table to record the basic information of the flight area, specifically including flight area id, flight area name, flight area boundary, flight area center point, etc.

[0049] Table 1. Flight area table LowFlightRegion

[0050]

[0051]

[0052] S12. Divide the flight area into multiple different tile-shaped airspace flight grids based on the CIM platform.

[0053] To achieve path planning for flight airspace in any range, the flight area is divided into multiple airspace flight grids of different tile blocks according to the geographical tile TMS (Tile Map Service) rule.

[0054] When dividing the airspace flight grid, it is divided according to the set geographical tile division rule and division range. Since the tiles are divided according to the quadtree, the division of the airspace flight grid can be achieved by recording the tile level and size.

[0055] S13. Use the GPU off-screen rendering technology to calculate the surface height of each airspace flight grid and store the calculated surface height.

[0056] The surface height usually refers to the vertical distance from a certain reference plane (water surface or selected mean sea level) to a certain point on the surface. This height includes the undulation of the terrain and the relative height of different ground objects (such as buildings, trees, etc.). The space above the surface height is considered as the area where low-altitude aircraft can fly.

[0057] Generally speaking, the surface height can be stored in the raster data in byte4 bytes. In a preferred embodiment, based on the off-screen rendering technology of the CIM platform, the surface height of each airspace flight grid is calculated, and the surface height data of each airspace flight grid is stored in the form of a picture (32-bit RGBA), that is, stored as a surface height map. The length and width of the picture are the width and height of the depth map, and the pixel value type of the picture is 32-bit double type.

[0058] The CPU depth buffer stores the distance from the camera to the target point, that is, the depth value. Through the inverse inference of the three-dimensional depth function, the surface height of each geographical target point can be calculated. Before performing GPU off-screen rendering, set the camera to orthographic projection and ensure that the camera position is directly facing the center point of the tile.

[0059] Preferably, in this embodiment, the tile blocks of the airspace flight grid after division in step S12 are three-dimensionally dissected into multiple spatial grids with a specification of 0.2 m × 0.2 m, so as to set the sampling resolution size of the surface height to 0.2 m × 0.2 m for calculating the width and height during off-screen rendering.

[0060] Figure 2 Schematically shows the result of obtaining the depth buffer by GPU off-screen rendering and performing visual rendering, and each pixel represents 0.2 m × 0.2 m of geography.

[0061] The formula for calculating the surface height corresponding to each pixel through the depth buffer is as follows:

[0062] Wp = Cp + Direction相机 *Distance 相机 - Object (1)

[0063]

[0064] Distance 相机-物体 = Near 相机 + Depth * (Far 相机 - Near 相机 ) (3)

[0065] where Wp = (X, Y, Z) represents the world coordinates of the object, the Z value represents the ground surface height, and Cp represents the world coordinates of the camera; Direction 相机 represents the camera orientation, and Distance 相机-物体 represents the distance from the camera to the object; Camera represents the camera center coordinates, Camera.X represents the camera longitude, Camera.Y represents the camera latitude, and Camera.Z represents the camera height; rowIndex and columnIndex represent the row and column numbers corresponding to the pixel; MapWidth and MapHeight respectively represent the length and width of the off-screen rendered image; Near 相机 represents the distance from the camera to the near plane, and Far 相机 represents the distance from the camera to the far plane; Depth is the depth buffer value obtained in the GPU, and its value range is (0 - 1).

[0066] S14. Establish a flight grid table and store the relevant data of the airspace flight grid in the flight grid table. Among them, the relevant data of the airspace flight grid include the id, name, ground surface height, geographical center position, geographical range bounding box of the airspace flight grid, and the flight area id corresponding to the airspace flight grid.

[0067] Among them, the number of airspace flight grids divided in step S12 is equal to the number of tiles obtained by dividing the flight area according to the geographical tile TMS. The suffix X_Y of the flight grid table name represents the row and column numbers corresponding to the geographical tile TMS. Preferably, the L17-level tile coding rule of the geographical tile TMS is adopted, and the length of each spatial grid is about 280 meters.

[0068] Table 2. Flight Grid Table LowFlightGrid_X_Y

[0069] Field Name Meaning Type id Airspace flight grid id, equivalent to gridId string regionId Corresponding flight area id string name Airspace flight grid name string geoLocation Geographical center position of the airspace flight grid string geoBoundingBox Geographical range bounding box of the airspace flight grid string rowIndex Row number in the case of TMS encoding int32 columnIndex Column number in the case of TMS encoding int32 depthImage Surface height map information of the airspace flight grid byte gridTmsLevel Level in TMS encoding int32 createTime Creation time datetime

[0070] S2. Set the flight path cost function; among all possible flight paths, select the path with the lowest cost as the planned path of the low-altitude aircraft.

[0071] To ensure the optimality and safety of the flight path of a low-altitude aircraft, each flight path is evaluated by calculating its flight cost. The flight cost is a dimensionless quantity. Among all possible flight paths, the path with the lowest cost is selected as the planned path for the low-altitude aircraft. The flight path cost function is as follows:

[0072] f cost = normalFly(d1) + clambFly(d2) + attractFly(d3) + repelFly(d4)

[0073] d = d1 + d2 + d3 + d4

[0074] Where d represents the flight distance, which is the sum of each segment of the flight trajectory in the flight path. The normalFly(d1) function represents the cost generated by the flight distance d1 of the low-altitude aircraft under the influence of external factors such as stable normal airflows. The clambFly(d2) function represents the cost generated by the flight distance d2 due to the influence of airflows during flight climb and flight above the standard flight altitude. At this time, compared with the cost per unit distance during stable normal airflows, the cost will increase, and this increase in cost also means an increase in the accident probability of the low-altitude aircraft caused by airflows. The attractFly(d3) function represents the cost generated by the flight distance d3 of the low-altitude aircraft when flying in the attraction area. At this time, compared with the cost per unit distance during stable normal airflows, the cost will decrease.

[0075] The repelFly(d4) function represents the cost generated by the flight distance d4 of the low-altitude aircraft when flying in the repulsion area. At this time, compared with the cost per unit distance during stable normal airflows, the cost will increase; for example, in a no-fly zone, the value of this function may become infinite, thus prompting the low-altitude aircraft to choose a detour; another example is that when the aircraft is too close to a building, a corresponding penalty buffer zone can be established based on factors such as privacy and safety, which will lead to an increase in flight cost.

[0076] Assume that the standard flight altitude of the low-altitude aircraft is H 标准飞行高度 , and the wind speed at the standard flight altitude H 标准飞行高度 is v0; assume that the wind speed at a certain flight altitude H 当前高度 is v1, then the relationship between the flight altitude and the wind speed is:

[0077]

[0078] In the formula, ρ is an empirical value set by the user, which is related to the atmospheric stability and ground roughness, and its value ranges from 0.125 to 0.5. The change of wind speed with flight altitude mainly depends on the ground roughness. Referring to Table 3, the ground roughness in urban areas is generally 0.4, and in this embodiment, the empirical value ρ is set to 0.4.

[0079] Table 3. Roughness value settings for different ground conditions

[0080] Ground condition ρ Smooth ground 0.1 There is more grass on the high ground in the city and very few trees 0.16 Many trees and very few buildings 0.22-0.24 Forest, village 0.28-0.30 There are high-rise buildings in the city 0.4

[0081] Generally, the flight cost of a low-altitude aircraft flying at a certain current altitude is as shown in the following function. For example, flying at an altitude of 80 meters compared to flying at an altitude of 60 meters, the flight cost increases by 12%.

[0082]

[0083] where d 飞行距离 is the flight distance at the current altitude.

[0084] S3. Plan the flight path.

[0085] The default flight mode in this embodiment is to first fly vertically upward to the standard flight altitude and then fly horizontally towards the destination. When encountering an obstacle during flight, it will fly around the side of the obstacle, or climb along the edge of the obstacle to a height exceeding the obstacle height and then fly horizontally. After flying over the obstacle, it will descend to the standard flight altitude and continue to fly horizontally until it reaches the designated destination point. To improve the efficiency of the flight path planning algorithm, this embodiment first plans a rough flight path, then designs a detour path based on the obstacles, and then optimizes the detour path to obtain the planned flight path.

[0086] S31. Initialize the flight path parameters.

[0087] Set parameters such as the starting and ending positions of the flight, the standard flight altitude, the maximum flight altitude limit, and the standard flight speed; according to the basic information of the flight area in step S1 (stored in the flight area table), initialize the flight path parameters.

[0088] S32. Perform three-dimensional dissection on the tile blocks of the airspace flight grid to obtain a three-dimensional space grid; based on the three-dimensional space grid and its grid height, generate a flight navigation map at the standard flight altitude according to the flight grid table; store the association relationship between the flight navigation map and the airspace flight grid as a two-dimensional array, and each value of the two-dimensional array is used to mark the spatial point type and data value of the spatial grid.

[0089] When flying at a certain standard flight altitude, it is necessary to first generate the flight navigation map at this standard flight altitude. The flight navigation map marks the flyable conditions of the flight area at a certain altitude.

[0090] When generating the flight navigation map, first perform three-dimensional dissection on the tile blocks of the airspace flight grid, divide them into spatial grids with a specification of 0.2 m × 0.2 m to obtain a three-dimensional space grid; after division, one spatial grid represents one spatial point.

[0091] Then, based on the three-dimensional space grid and its grid height, a flight navigation map is calculated according to the flight grid table established in step S1. Each flight navigation map corresponds to an airspace flight grid.

[0092] The association relationship between the flight navigation map and the airspace flight grid is stored as a two-dimensional array. The length and width dimensions of the two-dimensional array are equal to the length and width dimensions corresponding to the surface height map information of the airspace flight grid. The definition of the spatial point type is shown in Table 4 specifically. In Table 4, the spatial point type and data value of the spatial grid are marked by each value of the two-dimensional array, which is specifically expressed as follows:

[0093] flightCell[,] = [{type: string, value: float}]

[0094] In the two-dimensional array, type represents the flight condition type of the spatial point, and value corresponds to the value of the spatial grid. When the type is Free, it represents a free airspace and flight is possible, and the value of value is 1; when the type is Obstacles, it represents an obstacle area, and the low-altitude aircraft needs to climb or detour, and the value of value is the surface height information stored in the flight area table.

[0095] During the generation of the flight navigation map, more flight condition types can be dynamically superimposed according to the actual situation of the flight path. More spatial grid feature information refers to Table 4.

[0096] When the type is Attract, it represents an attraction area, and the value of value is the cost value of the attraction area; when the type is SafeBuff, it represents a safety buffer area, and the value of value is the cost value of the safety buffer area; when the type is Forbiden, it represents a no-fly area, and the value of value is 0.

[0097] Table 4. Definition of Spatial Point Types in Flight Navigation Map

[0098]

[0099] Among them, the spatial point type is an airspace control type, specifically including free airspace, attraction area, safety buffer area, no-fly area, etc. The no-fly area is a strictly prohibited flight area for low-altitude aircraft; the attraction area is to make the flight trajectory approach a safer area; the safety buffer area is a safety buffer distance set for low-altitude aircraft to avoid approaching obstacles.

[0100] During the flight, multiple airspace flight grids may be traversed. To improve efficiency, a fixed number of surface height map caches and flight navigation map caches are established to improve the algorithm efficiency that requires multiple database accesses and calculations during the flight path planning process.

[0101] Figure 3 It shows the visualization effect generated based on the flight navigation map. In the figure, the white area is the free area, and the black area represents the obstacle area.

[0102] S33. Initially plan the navigation path.

[0103] According to the flight navigation map at the standard flight altitude obtained in step S32, use the DDA algorithm to calculate the spatial points of the straight flight path between the starting point and the ending point. According to the superposition of the spatial points and the obstacles, divide the flight path into straight flight segments and obstacle segments. If a certain flight path segment has no obstacles, it is a straight flight segment, and set this flight path segment as part of the flight path. Otherwise, it is an obstacle section, and execute step S34.

[0104] Figure 4 It shows the initial planning of the navigation path, and according to the starting point and ending point of the flight path, breaks the initial flight path into a flight navigation map of obstacle sections and straight flight sections.

[0105] S34. Based on the bypass algorithm, select horizontal bypass or climbing bypass for the obstacle segment, and select the best bypass path according to the flight path cost to obtain the optimized flight path for each obstacle segment.

[0106] When a low-altitude aircraft encounters obstacles such as buildings at the standard flight altitude, it can choose to bypass. The bypass algorithm includes horizontal bypass and climbing bypass, and finally selects the best bypass path according to the flight path cost.

[0107] 1) Horizontal bypass

[0108] Horizontal bypass means flying along the edge of the obstacle. The bypass algorithm uses the Sobel graphic edge detection algorithm. As Figure 5 shown, the specific method is: Let the spatial grid point that starts to contact the obstacle be A, and the spatial grid point that the straight line leaves the obstacle be B. Starting from point A, perform a depth traversal of the surrounding spatial grid until the target point B for bypassing the obstacle is searched. During the search process, in order to closely bypass the obstacle, it is necessary to judge whether the spatial point to be accessed is the edge point of the obstacle during the traversal. Multiple bypass trajectories may be obtained during the search process, and calculate the flight cost α according to the formula in step S2, and select the path with the minimum cost. As Figure 6 、 Figure 7 shown.

[0109] 2) Climbing bypass

[0110] When the low-altitude aircraft does not exceed the maximum flight altitude limit, it can choose to climb and fly around. In this embodiment, the climbing flight strategy is set as flying vertically upward to a position exceeding the height of the obstacle first, then flying horizontally, and after flying over the obstacle, descending to continue flying horizontally.

[0111] As Figure 8 shown, when calculating the flight path around in the ascending direction, if the distance between the front and rear obstacles is less than the set threshold, the low-altitude aircraft does not need to descend after passing through the first obstacle and continues to fly. Conversely, it is necessary to descend after flying over the first obstacle, fly a certain distance, and then continue to climb to fly over the next obstacle.

[0112] 3) Select to fly around

[0113] Calculate and compare the flight costs of horizontal flight around and climbing flight around, and select the flight route with the least cost as the flight planning path.

[0114] S35. Merge the flight paths.

[0115] Based on the above steps, merge each direct flight segment and the optimized flight paths selected for each obstacle segment to obtain a complete planning path. The flight cost for this flight can also be further calculated as a reference cost.

[0116] S4. Optimize the flight path.

[0117] Since the path planned in step S3 is not optimal and there is much room for optimization. It is necessary to perform multiple rounds of simplification processing on the flight path to reduce unnecessary turns and detours. After optimizing the flight path, the flight cost should be reduced.

[0118] S41. Identify key spatial points.

[0119] Connect the spatial points on the flight path into lines, and simplify and merge the lines based on the line segment simplification algorithm, retaining the starting and ending points of the line segments, and finally obtaining the key spatial points in the flight path.

[0120] S42. Simplify the flight path by directly connecting based on any two key spatial points.

[0121] As Figure 9 shown, if there is no obstacle between a certain two key spatial points (i.e., key spatial grid points), it means that these two spatial points can fly directly, then delete the other spatial points between these two key spatial points. For example, spatial point 1 can be directly connected to spatial point n, making the flight distance shorter and the cost lower.

[0122] S43. Realize path simplification based on in-line interpolation points.

[0123] If the distance between two adjacent key spatial points is relatively long, after inserting interpolation points into their path, if the program of "direct connection simplification between any two key spatial points" can be implemented, then perform this operation, that is, directly connect and simplify the flight path based on any two interpolation points and / or key spatial points.

[0124] For example Figure 10 , spatial point 1 and spatial point 3 cannot be directly connected, but after inserting interpolation points, it can be directly connected to an interpolation point between spatial point 2 and spatial point 3.

[0125] S44. Optimize the wavy flight trajectory.

[0126] Based on the path planning algorithm generated in step S3, when encountering multiple groups of obstacles, the flight path will swing in a wavy manner around the center line, resulting in a longer flight distance. Therefore, it is necessary to perform dynamic planning on the trajectory path again, aiming to remove the path of flying back and forth. For example Figure 11 , the specific steps for re-planning the wavy flight trajectory are as follows:

[0127] (1) Identify the wave band

[0128] Rotate the path around the start and end points as the axis until the start and end points coincide with the x-axis. At this time, the positive wave band is above the x-axis and the negative wave band is below it.

[0129] (2) Simplify the wave crest trajectory

[0130] For example Figure 12 As shown, if the distance between two consecutive positive wave band wave crests exceeds the threshold, try to perform path planning on the two wave crest points according to step S3. If the distance of the re-planned path is less than the original path, update the path planning for this section. The same processing is also performed on the path of the negative wave band.

[0131] Further preferably, in the attraction area, the cost per unit flight distance will decrease. Therefore, if the airspace flight grid during the flight is close to the attraction area, perform a breadth-first traversal algorithm for the small-scale airspace flight grid, fly within the attraction area when approaching the airspace flight grid in the attraction area, and obtain a flight trajectory with a more optimal flight cost.

[0132] S45. Repeatedly execute the flight path optimization process of steps S41 - 44 until the change in flight cost is less than the preset threshold.

[0133] S5. Overlay the airspace control information and other temporary constraint conditions to obtain the final flight path.

[0134] The default flight path planning does not consider situations such as real airspace restrictions. Considering the diverse types, high update frequency, and strong flexibility of airspace control, the information of the airspace control area (i.e., airspace control information) is regarded as dynamic data to achieve efficient and flexible flight planning. The airspace control information mainly affects the content of the flight navigation chart, specifically including the airspace control type and the airspace control flexibility.

[0135] The airspace control types include no-fly zones, attraction zones, safety buffer zones, etc. A no-fly zone is an area where low-altitude aircraft are strictly prohibited from flying; an attraction zone is an area designed to make the flight trajectory approach a safer area; a safety buffer zone is a safety buffer distance set for low-altitude aircraft to avoid approaching obstacles.

[0136] The airspace control flexibility includes parameters such as the opening conditions, time conditions, flight costs, and aircraft type conditions of the airspace control area. It can specifically meet the airspace flight path planning under different conditions.

[0137] S51. Semantically express the airspace control information, obtain the semantic content of the airspace restrictions, and store it in the airspace control table.

[0138] The original data information of the airspace control area is vector data used to represent the range. To improve the calculation efficiency, the vector data of the airspace restricted area is cut according to the TMS slicing rule, rasterized, and then stored in the airspace control raster table, as Figure 13 shown.

[0139] The airspace control table is shown in Table 5. The semantic content of the airspace restrictions specifically includes the control area id, control area name, geometric information of the control area, geographical range bounding box of the control area, control area type, etc.

[0140] Table 5. Airspace Control Table LowFlightMask

[0141]

[0142] The airspace control raster table is shown in Table 6. It records the raster data after slicing the airspace control information according to the TMS rule, and also associates the spatial flight grid of the tile block with the airspace control information, that is, Table 6 reflects the association relationship between the flight grid table and the airspace control table.

[0143] Table 6. Airspace Control Raster Table LowFlightMaskRaster

[0144] Field Name Meaning Type id Associated id string maskId Corresponding special restricted area id string masktype Feature type string gridId Corresponding flight control grid Id string maskImage Convert the special restricted area into raster data byte

[0145] S52. Update the flight navigation chart during the flight path planning process.

[0146] In the flight navigation map generated in step S3, airspace control information and other temporary constraints are superimposed to obtain the final flight navigation map, such as Figure 14 .

[0147] Among them, other temporary constraints are mainly used for emergencies to avoid flight factors not considered in advance, so that the planned route can be adjusted in real time according to the actual situation. For example, an activity is temporarily organized in the current airspace area, but it was not known before the path planning. Therefore, this airspace area should be added as a temporary no-fly zone to the plan, and the flight navigation map is updated through this temporary no-fly zone.

[0148] Based on the same inventive concept, this embodiment also provides a low-altitude flight path planning system based on CIM, which is implemented based on the above path planning method. The path planning system includes:

[0149] A data fusion and storage module, which is used to divide the flight area based on the CIM platform to obtain multiple different airspace flight grids; calculate the surface height of each airspace flight grid and store the calculation results;

[0150] A flight cost calculation function, which is used to set the flight path cost function; among all possible flight paths, select the path with the lowest cost as the planned path of the low-altitude aircraft;

[0151] A path planning module, which is used to first plan a rough flight path, then design a detour path based on obstacles, and then optimize the detoured path to obtain the planned flight path;

[0152] A path optimization module, which is used to perform multiple rounds of simplification processing on the planned flight path to reduce unnecessary turns and detours and obtain the optimized flight path;

[0153] A condition overlay module, which is used to overlay airspace control information and other temporary constraints to obtain the final flight path.

[0154] In the path planning system of this embodiment, each module is respectively used to implement the corresponding steps of the foregoing path planning method. For the detailed implementation process, please refer to steps S1 - S5.

[0155] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present invention shall be equivalent replacement methods and are all included in the protection scope of the present invention.

Claims

1. A CIM-based low-altitude flight path planning method, characterized in that, It includes the following steps: S1. Divide the flight area based on the CIM platform to obtain multiple different airspace flight grids; calculate the surface height of each airspace flight grid and store the calculation results; S2. Set the flight path cost function; among all possible flight paths, select the path with the lowest cost as the planned path for the low-altitude aircraft; S3. First plan a rough flight path, then design a detour path based on obstacles, and then optimize the detoured path to obtain the planned flight path; S4. Perform multiple rounds of simplification processing on the flight path planned in step S3 to reduce unnecessary turns and detours and obtain the optimized flight path; S5. Overlay the airspace control information and other temporary constraint conditions to obtain the final flight path.

2. The low-altitude flight path planning method according to claim 1, wherein Step S1 includes: S11. Integrate multi-source data of the urban real scene model based on the CIM platform; establish a flight area table according to the integrated data to record the basic information of the flight area; S12. Divide the flight area into multiple different tile-based airspace flight grids based on the CIM platform; S13. Calculate the surface height of each airspace flight grid and store the calculated surface height; S14. Establish a flight grid table and store the relevant data of the airspace flight grid in the flight grid table; among them, the relevant data of the airspace flight grid includes the id, name, surface height, geographical center position, geographical range bounding box of the airspace flight grid, and the flight area id corresponding to the airspace flight grid.

3. The low-altitude flight path planning method according to claim 2, wherein The basic information of the flight area table includes the flight area id, flight area name, flight area boundary, and flight area center point; In step S12, when dividing the airspace flight grid, it is divided according to the set geographical tile division rules and division range.

4. The low-altitude flight path planning method according to claim 2, characterized in that, In step S13, the tiles of the airspace flight grid divided in step S12 are three-dimensionally dissected into multiple spatial grids, and then the GPU off-screen rendering technology is used to calculate the surface height of each airspace flight grid.

5. The low-altitude flight path planning method according to claim 2, wherein, Step S3 includes: S31. Set the flight start and end point positions, standard flight height, maximum flight altitude limit, and standard flight speed; initialize the flight path parameters according to the basic information of the flight area stored in the flight area table; S32. Three-dimensionally dissect the tiles of the airspace flight grid to obtain a three-dimensional spatial grid; based on the three-dimensional spatial grid and its grid height, generate a flight navigation map at the standard flight height according to the flight grid table; store the association relationship between the flight navigation map and the airspace flight grid as a two-dimensional array, and each value of the two-dimensional array is used to mark the spatial point type and data value of the spatial grid; the spatial point type is the airspace control type, including free airspace, attraction area, safety buffer zone, and no-fly zone; S33. Preliminarily plan the navigation path, calculate the spatial points of the straight flight path from the starting point to the end point according to the flight navigation map at the standard flight height; divide the flight path into straight flight segments and obstacle segments according to the superposition of the spatial points and obstacles; S34. Based on the detour algorithm, select horizontal detour or climb detour and select the best detour path according to the flight path cost; S35. Merge each direct flight segment and the optimized flight paths selected for each obstacle segment to obtain a complete planned path.

6. The low-altitude flight path planning method according to claim 5, wherein The airspace control information in step S5 is content that affects the flight navigation map, including the airspace control type and the airspace control flexibility; the airspace control flexibility includes the opening conditions, time conditions, flight costs, and aircraft type conditions of the airspace control area. Other temporary constraint conditions are used to avoid flight factors not considered in advance, so that the planned route can be adjusted in real time according to the actual situation.

7. The low-altitude flight path planning method according to claim 5, characterized in that Step S5 includes: S51. Semantically express the airspace control information, obtain the semantic content of the airspace restrictions and store it in the airspace control table; cut the vector data of the airspace restricted area according to the geographical tile cutting rules, rasterize it, and store it in the airspace control raster table. The semantic content of the airspace restrictions includes the control area id, the control area name, the geometric information of the control area, the geographical range bounding box of the control area, and the type of the control area. The airspace control raster table is used to record the raster data after the airspace control information is segmented into tiles according to the TMS rules, and is used to associate the spatial flight grid of the tile block and the airspace control information. S52. During the flight path planning process, superimpose the airspace control information and other temporary constraint conditions to update the flight navigation map.

8. The low-altitude flight path planning method according to claim 5, characterized in that, In a two-dimensional array, use type to represent the flight condition type of the spatial point, and use value to correspond to the value of the spatial grid. When the type is Free, it represents a free airspace and flight is allowed, and the value of value is 1; when the type is Obstacles, it represents an obstacle area, and low-altitude aircraft need to climb or detour, and the value of value is the surface height information stored in the flight area table; when the type is Attract, it represents an attraction area, and the value of value is the cost value of the attraction area; when the type is SafeBuff, it represents a safety buffer area, and the value of value is the cost value of the safety buffer area; when the type is Forbiden, it represents a no-fly area, and the value of value is 0.

9. The low-altitude flight path planning method according to claim 2, wherein Step S4 includes: S41. Connect the spatial points on the flight path into lines, and simplify and merge the lines based on the line simplification algorithm, retaining the start and end points of the lines, and finally obtaining the key spatial points in the flight path. S42. If there are no obstacles between the key spatial points, directly connect and simplify the flight path based on any two key spatial points to shorten the flight distance. S43. If the distance between the front and back key spatial points is long, insert interpolation points into the path between the front and back key spatial points, and then directly connect and simplify the flight path based on any two interpolation points and / or key spatial points. S44. Re-plan the wavy flight trajectory and remove the round-trip detour paths. S45. Repeat the flight path optimization process in steps S41 - 44 until the change in flight cost is less than the preset threshold.

10. The low-altitude flight path planning method according to claim 1, wherein The flight path cost function in step S2 is: f cost = normalFly(d1) + clambFly(d2) + attractFly(d3) + repelFly(d4) d = d1 + d2 + d3 + d4 where d represents the flight distance, which is the sum of each flight trajectory segment in the flight path. The normalFly(d1) function represents the cost generated by the flight distance d1 of a low-altitude aircraft under the influence of stable normal airflow factors; The clambFly(d2) function represents the cost generated by the flight distance d2 due to the influence of airflow during flight climb and flight above the standard flight altitude. At this time, the cost per unit distance of flight will increase compared to that under stable normal airflow; The attractFly(d3) function represents the cost generated by the flight distance d3 of a low-altitude aircraft when flying in the attraction area. At this time, the cost per unit distance of flight will decrease compared to that under stable normal airflow; The repelFly(d4) function represents the cost generated by the flight distance d4 of a low-altitude aircraft when flying in the repulsion area. At this time, the cost per unit distance of flight will increase compared to that under stable normal airflow.

11. A low-altitude flight path planning system based on CIM, characterized in that, Implemented based on the path planning method described in any one of claims 1-10, the path planning system includes: A data fusion and storage module, used to divide the flight area based on the CIM platform to obtain multiple different airspace flight grids; calculate the surface height of each airspace flight grid and store the calculation results; A flight cost calculation function, used to set the flight path cost function; select the path with the lowest cost as the planned path of the low-altitude aircraft among all possible flight paths; A path planning module, used to first plan a rough flight path, then design a detour path based on obstacles, and then optimize the detoured path to obtain the planned flight path; A path optimization module, used to perform multiple rounds of simplification processing on the planned flight path to reduce unnecessary turns and detours and obtain the optimized flight path; A condition overlay module, used to overlay airspace control information and other temporary constraint conditions to obtain the final flight path.

Images