A multi-unmanned aerial vehicle low-altitude coverage search task planning method in an urban environment

Abstract

The application discloses a kind of urban environment under multi-unmanned plane low-altitude coverage formula search task planning method, it is related to unmanned plane task planning field, comprising: according to the initial grid map of search area that unmanned plane constraint and map digital information establish;Full coverage scanning search is decomposed into ground scanning and building scanning, and according to the grid height relationship, map elements are classified into ground to be searched and scanned building;Calculate the area to be scanned of ground area and building surface, and complete the distribution of the number of unmanned plane with area ratio as distribution criterion;Ground scanning task distribution is realized based on the area iteration of subarea by establishing ground area division optimization model;Improved multi-traveler model is established, and building scanning task distribution is realized based on genetic algorithm solution.The application can realize the task planning of each unmanned plane in the multi-unmanned plane coverage reconnaissance task scene facing urban environment.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of unmanned aerial vehicle task planning, and particularly relates to a multi-unmanned aerial vehicle low-altitude coverage search task planning method in urban environment. BACKGROUND

[0002] As an economic and cultural center, the city has important strategic significance. However, the city environment is usually densely built and complex in terrain, and the traditional high-altitude photoelectric reconnaissance method is prone to information loss due to line-of-sight obstruction. Therefore, using small unmanned aerial vehicles for low-altitude omnidirectional detection in urban areas can significantly improve the comprehensiveness and accuracy of information collection, and has important application value.

[0003] Current research on urban target search task planning mainly focuses on the task planning of ground target search, and usually adopts a fixed altitude flight strategy for unmanned aerial vehicles, and task planning is carried out on this premise. However, this strategy does not fully consider the problem of building occlusion of ground targets at a specific angle, resulting in incomplete information collection in some areas. In addition, buildings in urban environments can serve as hiding places for target personnel, and their surfaces also need to be scanned. Therefore, a search strategy that only targets ground areas cannot meet the comprehensive search needs in complex urban environments. In summary, a new search strategy is needed that can fully consider the search needs of ground areas and building surfaces, and combine the advantages of multi-unmanned aerial vehicle cooperation to achieve efficient and comprehensive low-altitude coverage search tasks in urban environments. SUMMARY

[0004] The purpose of the present application is to provide a multi-unmanned aerial vehicle low-altitude coverage search task planning method in urban environment. Based on the scanning width required to ensure sensor accuracy, the length of the grid cell is established, the classification and extraction of ground and buildings are realized through the height relationship, the urban coverage search is divided into ground scanning and building scanning, and the number of unmanned aerial vehicles performing the corresponding tasks is allocated with the purpose of the shortest task time, and finally the task planning of each unmanned aerial vehicle is completed.

[0005] Technical scheme: In order to achieve the above purpose, the present application provides the following technical scheme:

[0006] In a first aspect, a multi-unmanned aerial vehicle low-altitude coverage search task planning method in urban environment is provided, comprising the following steps:

[0007] Establishing an initial grid map of the search area;

[0008] Decomposing the search task into ground scanning and building scanning, and dividing the grid in the initial grid map into two types of ground grid and building grid;

[0009] Based on the grid point partition result, a grid map for task planning is established;

[0010] The areas to be scanned for the ground scanning and the building scanning are calculated respectively, and the area ratio is taken as the allocation criterion to complete the number allocation of the unmanned aerial vehicles for the ground scanning and the building scanning.

[0011] Further, an initial grid map of the search area is established according to the unmanned aerial vehicle constraints and the map digital information, comprising:

[0012] Firstly, the DEM map and the DSM map of the same precision of the search area are acquired;

[0013] Then, the length of the grid unit in the initial grid map is calculated:

[0014] l X =l Y =Wη

[0015] wherein, l X and l Y are the horizontal and vertical lengths of the grid unit respectively, W is the scanning width of the sensor when the unmanned aerial vehicle flies at a height of h, and η is the lateral overlap rate during scanning;

[0016] Further, the size relationship between l X and l Y and the precision of the DEM and DSM maps is judged, if both l X and l Y are smaller than the precision of the DEM and DSM maps, the interpolation processing is performed on the DEM and DSM maps respectively, so that the distance between two adjacent points in the horizontal and vertical directions of the DEM and DSM maps is smaller than l X and l Y ;

[0017] Finally, the DEM height h E (i,j) and the DSM height h S (i,j) of each grid (i,j) in the initial grid map are calculated, and the calculation formulae are as follows:

[0018]

[0019] wherein, H E (x,y) and H S (x,y) are the DEM height and the DSM height of the point (x,y) respectively, and N is the number of points in the DEM and DSM maps that meet the constraints.

[0020] Further, the grid partition discriminant is as follows:

[0021]

[0022] wherein, hE (i,j) is the DEM height of the grid (i,j), h S (i,j) is the DSM height of the grid (i,j), Ah1 is the discrimination height, building is the building grid set, and ground is the ground grid set.

[0023] Further, based on the grid point division result, a grid map for task planning is established, including:

[0024] The height of all grids in ground is set to 0, and the height of the grid (i,j) corresponding to the grid in building is set to h(i,j) = h S (i,j) - h E (i,j) - Ah1.

[0025] The grids in building are classified to obtain a building set B = {B1,...,B k ,...,B M}, and the classification method is:

[0026]

[0027] wherein k = 1, 2,..., M, M is the total number of buildings, n k and m k are the number of grids horizontally contained and the number of grids vertically contained by the kth building in the initial grid map, respectively, and are the row index and the column index of the lower left corner grid of the building in the initial grid map, respectively, and k1 and k2 are classification coefficients;

[0028] Finally, the height of all grids in B k is updated to H k = max(h(i,j)), (i,j) e B k , to obtain the final grid map for task planning.

[0029] Further, the scanning areas to be scanned for ground scanning and building scanning are calculated respectively, and the area ratio is taken as the allocation criterion to complete the allocation of the number of unmanned aerial vehicles for ground scanning and building scanning, including:

[0030] The scanning area S G for ground scanning is calculated according to the number of ground grids, and the calculation formula is:

[0031] S G = Ml X l Y

[0032] where M is the number of grids in the ground set, l X and l Y are the horizontal and vertical lengths of the grid unit, respectively;

[0033] The scanned area S of the building is calculated according to the height relationship between the building grid and its adjacent peripheral grid. B The calculation formula is as follows:

[0034]

[0035] where S is the scanned area of the building B k .

[0036] Further, the number of UAVs for ground scanning and building scanning is allocated according to the area ratio, including:

[0037] The number of UAVs for ground scanning N GU and the number of UAVs for building scanning N BU are respectively represented as:

[0038]

[0039] N BU = N U -N GU

[0040] where N U is the total number of UAVs, and represent the floor and ceiling functions, respectively; ε is the dispersion coefficient of the building, satisfying ε>1.

[0041] Further, the method further comprises:

[0042] The allocation of the ground scanning task is iteratively realized based on the sub-area area, including:

[0043] Step 1: According to the number of UAVs for ground scanning N GU , the area covered by the ground grid set ground is initially divided to obtain a sub-area set where each sub-area is rectangular and there is no overlap between any two sub-areas, q=1, 2, …, N GU .

[0044] Step 2: Calculate the sum T q of the time for the UAV U q to fly to the center of the sub-area Q q and the time for scanning the sub-area Q q , the calculation formula is as follows:

[0045]

[0046] Where V is the speed of the drone, L is q For U q Starting point distance Q q Euclidean distance of the center point, S q Q q area;

[0047] Step 3: Get the time set Find max(T) and min(T) and the corresponding sub-area index respectively max and index min ;

[0048] Step 4: Calculate the difference Δt = max(T) - min(T). If Δt is less than the preset value or the number of iterations is greater than the maximum number of iterations, terminate and output the current sub-region set.

[0049] Step 5: Follow the steps from index min To index max The transmission direction of the sub-area index min and sub-regions of the pathway (excluding index max ) adds a row or a column outward along one edge of the y-axis to obtain a new set of sub-regions and return to Step 2.

[0050] Furthermore, the method further comprises:

[0051] An improved multi-traveling salesman model is established and solved based on a genetic algorithm to achieve the allocation of building scanning tasks, including:

[0052] Build an improved multiple traveling salesman model:

[0053]

[0054] Among them, seq i For UAV U i The task sequence, ind v Refers to seq i The vth element represents the number of the building in the building set B. seq i | is seq i The number of elements in

[0055] Buildings and Flight time between As edge weights, buildings Scan time As the node weight, f(seq i ) is the calculation formula:

[0056]

[0057] In the formula, l k,k+1 is the distance between the building and , V is the flight speed, is the area to be scanned of the building ;

[0058] The task sequence is taken as the gene code to obtain the task sequence of the unmanned plane scanning the building by the genetic algorithm.

[0059] The second aspect also provides a computer readable storage medium storing one or more programs, the one or more programs including instructions that, when executed by a computing device, cause the computing device to perform the multi-unmanned plane low-altitude coverage search task planning method in a city environment as described above.

[0060] The third aspect also provides an electronic device including one or more processors, one or more memories, and one or more programs, wherein the one or more programs are stored in the one or more memories and configured to be executed by the one or more processors, and the one or more programs include instructions for performing the multi-unmanned plane low-altitude coverage search task planning method in a city environment as described above.

[0061] Beneficial effects: the multi-unmanned plane low-altitude coverage search task planning method in a city environment provided by the present application fully considers the establishment of the grid map based on the scanning width constraint, and realizes the rapid decomposition of the ground and the building according to the grid map, the method considers the relationship between the flight height of the unmanned plane and the sensor accuracy, so that the unmanned plane can maintain the same flight height to the ground in the whole task process; the strategy that the ground and the building are searched and scanned by two different teams of unmanned planes is designed, the original coupling problem is processed in layers, and the difficulty of problem solving is reduced; and the unmanned plane number allocation method and the task execution timing optimization model are designed based on this, so that the task time is the shortest. Based on the present application, the coverage scanning task of the unmanned plane formation can be quickly distributed. BRIEF DESCRIPTION OF DRAWINGS

[0062] Figure 1 is the multi-unmanned plane low-altitude coverage search task planning flowchart for a city environment provided by the present application.

[0063] Figure 2 is the DEM schematic diagram used in the present example.

[0064] Figure 3is a DSM diagram used in the present example.

[0065] Figure 4 is a ground and building distribution diagram obtained by using the method in the present example.

[0066] Figure 5 is a final grid map for task planning obtained by using the method in the present example.

[0067] Figure 6 is a passing direction.

[0068] Figure 7 is a flowchart based on sub-area area iteration provided by the present application.

[0069] Figure 8 is a sub-area division result diagram obtained by using the method in the present example. DETAILED DESCRIPTION

[0070] In order to make the content of the present application more easily understood, the present application will be further described in detail below in combination with specific embodiments. In the present example, four unmanned aerial vehicles search a city area.

[0071] As shown in Figure 1 , the present embodiment provides a multi-unmanned aerial vehicle low-altitude coverage search task planning method in a city environment, which comprises the following steps:

[0072] Step 1: establishing a grid map according to unmanned aerial vehicle constraints and map digital information.

[0073] Firstly, a DEM map and a DSM map of the same precision of the search area are obtained. In the present example, the obtained DEM map and DSM map are as shown in Figure 2 、 Figure 3 , which are maps converted from longitude and latitude to length, wherein the unit length of the z-axis is m, the unit length of the x-axis and y-axis is 8 m, and each map has 60x60=3600 data points.

[0074] Then, the length of the grid unit in the grid map is calculated, and the calculation relationship is as follows:

[0075] l X =l Y =Wη

[0076] wherein l X and l Y are the horizontal and vertical axis lengths of the grid unit, W is the scanning width of the sensor when the unmanned aerial vehicle flies at a height of h, η is the lateral overlap rate when scanning, W is 20 meters, and η is 80%, so that l X =l Y =16 meters. Since l X =lY = 16 greater than 8 meters, no need to interpolate DEM and DSM map.

[0077] Finally, based on the length of the grid cell to establish a grid map, the DEM height h E (i,j) and DSM height h S (i,j) are calculated, and the calculation formula is respectively:

[0078]

[0079] Where H E (x,y) and H S (x,y) are the DEM height and DSM height corresponding to the point (x,y) respectively, and N is the number of points in the DEM map and the DSM map that meet the constraints, N = 4.

[0080] Step two: the scanning search is disassembled into ground scanning and building scanning, and the grid in the grid map is divided into ground grid and building grid according to the height relationship of the grid, and then the map is divided into ground to be searched and building to be scanned, which specifically includes:

[0081] 1. In this embodiment, the scanning tasks of the ground area and the building area are respectively assigned to different unmanned aerial vehicle formations for execution; this strategy divides the original integrated coupled solving problem into layers, reducing the complexity of problem solving.

[0082] 2. According to the size relationship between h E (i,j) and h S (i,j), building grid and ground grid are generated, and the discriminant is:

[0083]

[0084] Where Δh1 is the discriminant height, building refers to the building grid set to be scanned, and ground refers to the ground grid set to be scanned, and in this example Δh1 = 10.

[0085] 3. In the grid map, set all the grid heights in ground to 0, and set the grid height corresponding to the grid (i,j) in building to h S (i,j) = h E (i,j) - Δh1. Figure 4 The figure shows the distribution of ground and building, where the gray grid is the building grid and the white grid is the ground grid.

[0086] 4. Classify the grids in building to obtain the building set B = {B1,...,B k ,...,BM} is the total number of buildings, and the classification method is:

[0087]

[0088] where n k and m k are the horizontal length and vertical length of building B k , respectively. and are the horizontal length and vertical length of building B k in the grid map, respectively.

[0089] 5. Let L k , W k and H k be the corresponding horizontal length, vertical length and height of B k , respectively, and their calculation formulas are:

[0090]

[0091] 6. Update the height of all grids in B k to H k , and obtain the final grid map for task planning as shown in FIG. 6. Table 1 shows the length, width and height of each building in this embodiment. Figure 5 Table 1 Length, width and height of buildings

[0092]

[0093] Serial number L / m W / m H / m 1 48 64 81 2 64 80 70 3 48 64 56 4 48 96 70 5 64 32 8 6 64 80 44 7 80 64 24 8 48 64 26.5 9 64 48 20 10 90 48 69

[0094] Step 3: Calculate the area to be scanned of the ground area and the building surface, and complete the distribution of the number of unmanned aerial vehicles by taking the area ratio as the distribution criterion. Specifically:

[0095] 1. Calculate the area to be scanned S G of the ground area according to the number of ground grids, and the calculation formula is:

[0096] S G = Ml X l Y

[0097] where M is the number of grids in the ground set, and M = 749 in this embodiment, so S G = 191744 m 2 .

[0098] ​2. The building to be scanned surface area is only its side surface, not including the top and bottom surface. Considering the case of two adjacent buildings, their coincident edges will reduce the surface area to be scanned, so the building surface area to be scanned cannot be obtained by simply adding all the building surfaces. The building surface area to be scanned is calculated according to the height relationship between the building outer grid and its adjacent peripheral grid, and its discrete calculation formula is as follows:

[0099]

[0100] wherein is the surface area to be scanned of the building B k , and S B = 118652 m 2 .

[0101] 3. Since the total number of UAVs is N U = 4, the number of UAVs N GU , N BU performing ground scanning and building scanning can be represented as:

[0102]

[0103] N BU = N U -N GU

[0104] wherein, and represent the floor and ceiling functions respectively, and ε is the dispersion coefficient of the building, in this example, ε = 2, so N GU = 2, N BU = 2.

[0105] Step four: iterative implementation of ground scanning task allocation based on sub-region area; specifically including:

[0106] Step 1: first, according to the number of UAVs N GU scanning the ground, the total area is initially divided to obtain a set of sub-regions Q Each sub-region Q q should be a regular rectangle, and there is no overlap between sub-regions;

[0107] Step 2: calculate the sum T q of the time for UAV U q to fly to the center of sub-region Q q and the time for scanning sub-region Q q , the calculation formula is:

[0108]

[0109] In the formula, V is the speed of the UAV, L q is the distance from the center of sub-region Q q to the center of the next sub-region Q q , and T q is the time for the UAV to fly to the center of the next sub-region Q q .q Starting point distance Q q Euclidean distance of the center point, S q Q q The area to be scanned.

[0110] Step 3: Get the time set Find max(T) and min(T) and the corresponding sub-area index respectively max and index min ;

[0111] Step 4: Calculate the difference Δt = max(T) - min(T). If Δt is less than the preset value or the number of iterations is greater than the maximum number of iterations, terminate and output the current sub-region set.

[0112] Step 5: Follow the steps below: Figure 6 As shown from index min To index max The transmission direction of the sub-area index min and sub-regions of the pathway (excluding index max ) adds a row or a column outward along one edge of the y-axis to obtain a new set of sub-regions and return to Step 2.

[0113] In order to explain the above process more clearly, the flow chart is as follows Figure 7 shown.

[0114] In this example, the starting point of each drone is the lower left corner of the map, the flight speed is 5m / s, the maximum number of iterations is 10, the initial division is equal to the left and right, and the difference preset value is 1min. The sub-area distribution corresponding to drone 1 and drone 2 can be obtained as follows Figure 8 As shown, at this time Δt = 48.3 seconds.

[0115] Step 5: Establish an improved multi-traveling salesman model and solve it based on the genetic algorithm to realize the allocation of building scanning tasks; specifically:

[0116] Build an improved multiple traveling salesman model:

[0117]

[0118] Among them, seq i Refers to the mission sequence containing the building number, corresponding to the UAV U i Done. seq i Expressed as ind v Refers to seq i The vth element represents the number of the building in the building set B. seqi The number of middle elements.

[0119] The flight time between the buildings And The scanning time of the building As the edge weight, the flight time between the buildings The scanning time of the building As the node weight, the f(seq i ) calculation formula is:

[0120]

[0121] In the formula, l k,k+1 The distance between the buildings And The flight speed, The area to be scanned of the building .

[0122] Finally, the task sequence is used as the gene code to perform the genetic algorithm to obtain the task sequence of the unmanned aerial vehicle scanning the building, that is, the order of the building scanned by each unmanned aerial vehicle performing the search task.

[0123] In this example, the number of iterations of the genetic algorithm is set to 5000, the initial population size is 100, the crossover probability is 0.4, and the mutation probability is 0.1, and the task sequences of the unmanned aerial vehicle 3 and the unmanned aerial vehicle 4 are {5, 10, 3, 7, 4} and {9, 8, 1, 6, 2} respectively, at this time, the time difference of the two unmanned aerial vehicles completing the task is 18.5 seconds.

[0124] Obviously, the above embodiments are only examples for clearly illustrating, and not limited to the use. For those skilled in the art, other different forms of changes or variations can be made on the basis of the above description. Here, all the use modes do not need to be exhausted, and the obvious changes or variations derived therefrom are still within the protection scope of the present application.

[0125] Based on the same technical solution, the application also discloses a computer readable storage medium storing one or more programs, the one or more programs comprising instructions which, when executed by a computing device, cause the computing device to perform the above-mentioned multi-unmanned aerial vehicle low-altitude coverage search route planning method in urban environment.

[0126] Based on the same technical solutions, the application further discloses a computing device, comprising one or more processors, one or more memories, and one or more programs, wherein the one or more programs are stored in the one or more memories and configured to be executed by the one or more processors, and the one or more programs comprise instructions for executing the method for planning a search flight path of multiple unmanned aerial vehicles in a city environment.

[0127] Those skilled in the art will understand that embodiments of the application can be provided as methods, systems, or computer program products. Therefore, the application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0128] The application is described with reference to the flowcharts and / or block diagrams of the methods, apparatus (systems), and computer program products according to embodiments of the application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus generate a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus for performing the functions specified in the flowcharts and / or block diagrams.

[0129] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including instruction apparatus, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus for performing the functions specified in the flowcharts and / or block diagrams.

[0130] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a process for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus for performing the functions specified in the flowcharts and / or block diagrams.

Claims

1. A method for multi-UAV low-altitude coverage search task planning in an urban environment, characterized in that, The method comprises: establishing an initial grid map of a search area; decomposing a search task into ground scanning and building scanning, and dividing the initial grid map into two types of ground grids and building grids; the division criterion of the grids is: wherein h E (i,j) is the DEM height of the grid (i,j), h S (i,j) is the DSM height of the grid (i,j), Δhi is the discrimination height, building is the building grid set, and ground is the ground grid set. establishing a grid map for task planning based on the grid point division result; calculating the scanning areas of the ground scanning and the building scanning respectively, and completing the distribution of the number of unmanned aerial vehicles for the ground scanning and the building scanning by taking the area ratio as the distribution criterion; The method further comprises iteratively implementing the distribution of the ground scanning task based on the sub-area area: Step1: According to the number of unmanned aerial vehicles N scanned on the ground GU , the initial division of the region covered by the ground grid set ground obtains a sub-region set , wherein each sub-region is rectangular and there is no overlap between any two sub-regions, q = 1, 2, …, N GU ; Step 2: Calculate the time T of the UAV U q flying to the sub-region Q q The sum of the time and the scanning sub-region Q q of the center T q , the calculation formula is: In the formula, V is the speed of the UAV, L q is the distance from the starting point position Q q to the center point S q , and S q is the Euclidean distance from the center point to the starting point position Q q . Step3: get the time set find max(T) and min(T) and the corresponding sub-region index index max and index min ; Step 4: Calculate the difference Δt = max(T) - min(T), if Δt is less than a preset value or the number of iterations is greater than the maximum number of iterations, terminate and output the current sub-area set; Step5: According to the transfer direction from index min to index max , increase a row or a column to the edge of the sub-region index min and the sub-region of the path, get a new sub-region set and return to Step2; wherein the sub-region of the path does not contain index max .

2. The method of claim 1, wherein, establishing an initial grid map of a search area according to the unmanned aerial vehicle constraints and the map digital information, comprising: Firstly, obtain the DEM map and the DSM map of the search area with the same precision; Then, calculate the length of the grid unit in the initial grid map: l X = l Y = Wη wherein, l X and l Y are the horizontal and vertical lengths of the grid cell, W is the scan width of the sensor when the UAV is flying at a height of h above the ground, and η is the lateral overlap ratio when scanning. Then, judge l X and l Y The size relationship with DEM, DSM map precision, if l X and l Y Both less than DEM, DSM map precision, respectively to DEM map and DSM map interpolation processing, make DEM, DSM map between the horizontal and vertical adjacent two points distance less than l X and l Y ; Finally, the DEM height h for each grid (i,j) in the initial raster map is calculated E (i,j) and the DSM height h S (i,j) is calculated, with the formulas respectively: where H E (x,y) and H S (x,y) are the DEM height and the DSM height corresponding to the point (x,y), respectively, and N is the number of points in the DEM map and the DSM map that satisfy the constraint.

3. The method of claim 1, wherein, establishing a grid map for task planning based on the grid point division result, comprising: Set all grid heights in ground to 0, and set the grid height for grid (i,j) in building to h(i,j) = h S (i,j) - h E (i,j) - Δh1; The grids in the building are classified to obtain a building set B = {B1,..., B k ,...,B M}, and the classification method is: wherein k = 1, 2,..., M, M is the total number of buildings, n k and m k are the number of grids contained in the horizontal direction and the number of grids contained in the vertical direction of the kth building in the initial grid map, respectively, and are the row index and the column index of the lower-left grid of the building in the initial grid map, respectively, and k1 and k2 are classification coefficients. Finally, update the grid height of all grids corresponding to B k to H k = max(h(i,j)), (i,j) ∈ B k , and get the final grid map for task planning.

4. The method of claim 1, wherein, calculating the scanning areas of the ground scanning and the building scanning respectively, and completing the distribution of the number of unmanned aerial vehicles for the ground scanning and the building scanning by taking the area ratio as the distribution criterion, comprising: S = (N - 1) * 2 * 2 * 2 G S = (N - 1) * 2 * 2 * 2 S G = Ml X l Y where M is the number of grids in the ground set, l X and l Y are the horizontal and vertical lengths of the grid cell, respectively. The scanned surface area S to be scanned of the building scan is calculated in relation to the height of the building outer grid and its adjacent peripheral grid B with the formula: In the formula is the area to be scanned of the building B k .

5. The method of claim 4, wherein, completing the distribution of the number of unmanned aerial vehicles for the ground scanning and the building scanning by taking the area ratio as the distribution criterion, comprising: The number of drones N scanning the ground GU and the number of drones N scanning the building BU are respectively expressed as: N BU = N U - N GU where N U is the total number of UAVs, and denote the floor and ceiling functions, respectively; and ε is the dispersion coefficient of the buildings, satisfying ε>

1.

6. The method of claim 1, wherein, The method further comprises: establishing an improved multi-traveler model and solving it based on a genetic algorithm to implement the distribution of the building scanning task, comprising: establishing an improved multi-traveler model: wherein seq i is a task sequence of the UAV U i , ind v is a number of the building represented by the vth element of seq i in the building set B, |seq i | is a number of elements in seq i . between buildings and the time of flight between buildings as edge weights, and the scan time of a building i as node weights, then f(seq i ) is calculated as wherein l k,k+1 is the distance between the building and the flying speed, is the area to be scanned of the building . taking the task sequence as the gene code to solve the genetic algorithm to obtain the task sequence of the unmanned aerial vehicle scanning the building.

7. A computer-readable storage medium storing one or more programs, the one or more programs comprising instructions that are executable to: The instructions, when executed by a computing device, cause the computing device to perform the method of any one of claims 1-6.

8. An electronic device, comprising: Comprising one or more processors, one or more memories, and one or more programs, wherein the one or more programs are stored in the one or more memories and configured to be executed by the one or more processors, the one or more programs comprising instructions for performing the method of any one of claims 1-6.

Images