Fast target assignment method suitable for large-scale unmanned vehicle group and related equipment

Abstract

The application discloses a kind of fast target allocation method suitable for large-scale unmanned aerial vehicle group and related equipment, it is related to unmanned aerial vehicle cluster control technical field.The application is introduced by introducing auction algorithm, using the parallel bidding mechanism of auction algorithm, so that multiple unmanned aerial vehicles independently execute target search and bidding decision each time, the parallel characteristics of auction algorithm are fully exerted, the matching convergence speed is greatly accelerated, the matching time is greatly reduced, and the fast target matching of large-scale unmanned aerial vehicle group is realized;In addition, through the cyclic matching mode, the solution speed is accelerated while the allocation quality is improved, so that the final matching result is basically consistent with the optimal matching result.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle cluster control, and particularly relates to a rapid target assignment method suitable for large-scale unmanned aerial vehicle groups and related equipment. BACKGROUND

[0002] In recent years, unmanned aerial vehicle light show performances have been widely used in large-scale celebrations, commercial promotion, cultural tourism and other fields due to their dynamic programmability, spatial flexibility and stunning visual effects. Unmanned aerial vehicle light show performances require a large number of unmanned aerial vehicles to move safely and efficiently in three-dimensional space in coordination to accurately achieve the expected pattern. The technical core lies in solving the path planning problem of large-scale unmanned aerial vehicle groups. Among them, target assignment is the first step of unmanned aerial vehicle group path planning, and the assignment efficiency and assignment quality will directly affect the efficiency of the overall path planning and the smoothness of the performance.

[0003] Currently, the target assignment methods used in unmanned aerial vehicle performances on the market generally use a target assignment algorithm based on the Hungarian algorithm. Although this target assignment algorithm can obtain a globally optimal task-unmanned aerial vehicle matching result, it needs to repeatedly perform matrix row and column operations, and the time complexity is O(n³). The algorithm complexity increases cubically with the number of unmanned aerial vehicles, and parallel computing cannot be performed. When the size of the unmanned aerial vehicle performance increases (such as more than 5000), the matching time of this target assignment algorithm is too long, greatly increasing the overall path planning time of the unmanned aerial vehicle group, and seriously restricting the real-time performance of large-scale performances.

[0004] In summary, the target assignment method used in existing unmanned aerial vehicle performances has the problem of too long matching time when processing the target assignment task of large-scale unmanned aerial vehicle groups, which seriously affects the overall path planning efficiency. SUMMARY

[0005] The technical problem to be solved by the present application is to provide a rapid target assignment method suitable for large-scale unmanned aerial vehicle groups and related equipment to solve the technical problem of too long matching time when processing the target assignment task of large-scale unmanned aerial vehicle groups by the target assignment method used in existing unmanned aerial vehicle performances.

[0006] To solve the above technical problems, the present application adopts the technical solutions as follows:

[0007] The first aspect of the present application provides a rapid target assignment method suitable for large-scale unmanned aerial vehicle group, comprising the following steps: S10, constructing a mathematical model of unmanned aerial vehicle group target matching problem based on auction algorithm; S20, initializing a to-be-assigned unmanned aerial vehicle set U, a target point cost set P, an assignment set A, an increment parameter epsilon, an increment scaling ratio k and a task benefit of the target point; S30, solving the mathematical model based on the auction algorithm to perform bidding matching for each unmanned aerial vehicle, and updating the to-be-assigned unmanned aerial vehicle set U, the target point cost set P and the assignment set A based on the matching result; S40, judging whether the increment parameter epsilon is greater than 1, if yes, executing step S50; if no, returning to the assignment set A; S50, updating the increment parameter epsilon=epsilon / k, retaining the current cost of each target point, resetting the to-be-assigned unmanned aerial vehicle set U and the assignment set A; S60, solving the mathematical model based on the auction algorithm to perform bidding matching for each unmanned aerial vehicle, updating the to-be-assigned unmanned aerial vehicle set U, the target point cost set P and the assignment set A based on the matching result, and jumping to step S40.

[0008] The second aspect of the present application provides a rapid target assignment system suitable for large-scale unmanned aerial vehicle group, comprising at least one processor and at least one memory connected in communication with the processor, wherein the memory stores program instructions, and the program instructions are called and executed by the processor to realize the rapid target assignment method suitable for large-scale unmanned aerial vehicle group.

[0009] The third aspect of the present application provides a computer readable storage medium, which internally stores program instructions, and the program instructions are executed by the processor to realize the rapid target assignment method suitable for large-scale unmanned aerial vehicle group.

[0010] The beneficial technical effects of the present application are that: by introducing the auction algorithm, the parallel bidding mechanism of the auction algorithm is used, so that multiple unmanned aerial vehicles independently perform target search and bidding decision each time, the parallel characteristics of the auction algorithm are fully utilized, the matching convergence speed is greatly accelerated, the matching time is greatly reduced, and the rapid target matching of large-scale unmanned aerial vehicle group is realized; in addition, through the cyclic matching mode, the solving speed is accelerated, and the assignment quality is improved, so that the final matching result is basically consistent with the optimal matching result. BRIEF DESCRIPTION OF DRAWINGS

[0011] Figure 1 It is a flowchart of the rapid target assignment method suitable for large-scale unmanned aerial vehicle group of the present application;

[0012] Figure 2 It is a schematic diagram of the comparison test results of 11800 times;

[0013] Figure 3 It is a ratio schematic diagram of the comparison test results of 11800 times;

[0014] Figure 4 The relative error diagram of 11800 flight times of comparative test results. DETAILED DESCRIPTION

[0015] In order to make the ordinary skilled in the art more clearly understand the purpose, technical scheme and advantages of the present application, the present application is further described below in conjunction with the drawings and examples.

[0016] As Figure 1 shown, in one embodiment of the present application, the rapid target assignment method suitable for large-scale UAV group mainly includes steps S10 to S60:

[0017] S10, constructing a mathematical model of UAV group target matching problem based on auction algorithm.

[0018] For the original UAV group target matching problem, it can be modeled as the following mathematical problem:

[0019] ;

[0020] Wherein, is the benefit obtained by the UAV i to the target point j, the matrix X represents the assignment matrix composed of decision variable elements , represents that the target point j is assigned to the UAV i, and represents that the target point and the UAV maintain one-to-one correspondence. When the total benefit of the assignment objective function is maximum, the matching result obtained is the optimal matching.

[0021] Based on the duality principle and optimality principle, the above mathematical problem can be transformed into:

[0022] ;

[0023] And is the new variable to be solved after transformation, wherein represents the current cost of the target point j, represents the benefit obtained by the UAV i. By observing the dual problem, it is found that the feasible solution of the mathematical problem must satisfy the constraint: , which is equivalent to . Therefore, according to the dual problem, the above mathematical optimization can be further optimized as:

[0024] ;

[0025] Since then, the original UAV group target matching problem is optimized as a mathematical model based on auction algorithm, wherein P is a target point cost set, including the current cost of all target points; represents the task benefit of target point j to UAV i; represents the current cost of target point j, and n is the number of UAVs.

[0026] By solving the mathematical model based on the auction algorithm, the UAV group target matching result can be determined. The mathematical model based on the auction algorithm can be understood as that when any UAV i gets the target position j with the highest benefit to itself at a lower price, the expected balance is reached, that is, the optimal solution is obtained.

[0027] S20, initializing the UAV set to be allocated U, the target point cost set P, the allocation set A, the increment parameter ε, the increment scaling ratio k, and the task benefit of the target point.

[0028] Step S20 is an initialization step, and the specific process includes:

[0029] S21, obtaining UAV information, including UAV id information and UAV position information.

[0030] Each UAV contains two pieces of information: id information and position information (wherein the id information refers to the number of each UAV in the UAV group), and the two pieces of information are stored using a structure D represents the id information of UAV i, represents the position information of UAV i.

[0031] S22, constructing and initializing the UAV set to be allocated U, and the initialized UAV set to be allocated U includes id information of all UAVs.

[0032] The UAV set to be allocated U constructed in this step is used to store the id information of all UAVs that have not been allocated target points. The initialized UAV set to be allocated U contains id information of all UAVs (the total number of UAVs is n), that is:

[0033] .

[0034] S23, constructing and initializing the target point cost set P, and the cost value of all target points in the initialized target point cost set P is 0.

[0035] The target point cost set P constructed in this step is used to store the current cost of all target points:

[0036] ,

[0037] wherein represents the current cost of target point j, The value of the target point cost set P is initialized as 0, i.e. at the initial moment .

[0038] S24, a distribution set A is constructed and initialized, and the initialized distribution set A is an empty set.

[0039] The distribution set A constructed in this step is used to store the current target distribution. The initialized distribution set A is an empty set, i.e. .

[0040] S25, set the increment parameter ε.

[0041] In this step, the increment parameter ε is initially set as ε > n / 10, n being the number of UAVs.

[0042] S26, set the increment scaling ratio k, k > 1.

[0043] S27, initialize the task revenue of each target point.

[0044] In this step, the task revenue of each target point is calculated using the following formula:

[0045] ,

[0046] wherein, represents the task revenue of the target point j to the UAV i, represents the position information of the UAV i, represents the position information of the target point j.

[0047] S30, the mathematical model based on the auction algorithm is solved using the auction algorithm to perform bidding matching for each UAV, and the to-be-distributed UAV set U, the target point cost set P, and the distribution set A are updated based on the matching result.

[0048] The bidding matching for each UAV includes the following steps:

[0049] S31, select P UAVs from the to-be-distributed UAV set U;

[0050] S32, perform bidding matching in parallel for the selected P UAVs using the auction algorithm, and update the to-be-distributed UAV set U, the target point cost set P, and the distribution set A based on the matching result.

[0051] In step S32, the selected P UAVs perform operations in parallel:

[0052] S321, find the target point j with the highest revenue to the UAV i, and calculate the highest revenue:

[0053] ,

[0054] wherein, represents the task benefit of the target point j to the unmanned aerial vehicle i, represents the current cost of the target point j.

[0055] S322, find the target point m with the second highest benefit to the unmanned aerial vehicle i, and calculate the second highest benefit:

[0056] ,

[0057] wherein, represents the task benefit of the target point m to the unmanned aerial vehicle i, represents the current cost of the target point m.

[0058] S323, the unmanned aerial vehicle i bids for the target point j:

[0059] ,

[0060] wherein, represents the current cost of the target point j, is the highest benefit to be found, is the second highest benefit to be found, is a set incremental parameter.

[0061] S324, the target point j is assigned to the unmanned aerial vehicle i, and the assignment is stored in the assignment set A, and the unmanned aerial vehicle is removed from the unmanned aerial vehicle set U to be assigned.

[0062] In the step S324, when the target point j is assigned to the unmanned aerial vehicle i, if the target point j has been assigned to other unmanned aerial vehicles, the previous assignment is removed, and the unmanned aerial vehicle previously assigned to the target point j is moved from the assignment set A to the unmanned aerial vehicle set U to be assigned.

[0063] S325, update the cost value of the target point j:

[0064] ,

[0065] wherein, is the bid price of the unmanned aerial vehicle i for the target point j.

[0066] The step S32 is to simultaneously perform bidding matching for P unmanned aerial vehicles, and for steps S321 to S325, during the period from the step S323 in which the unmanned aerial vehicle i bids for the target point j to the step S324 in which the target point j is assigned to the unmanned aerial vehicle i, the target point j is locked by means of a global lock, so as to ensure that data does not conflict during parallel execution.

[0067] After the P drones to be selected complete the bidding matching through steps S321 to S325, it is judged whether the set of drones to be allocated U is empty, if yes, the iteration is ended, if not, it is jumped to step S31 to continue iteration, that is, steps S31 to S32 are repeatedly executed until the set of drones to be allocated U is empty.

[0068] S40, it is judged whether the increment parameter ε is greater than 1, if yes, step S50 is executed, if not, the set A of allocation is returned.

[0069] It should be noted that, based on the principle of the auction algorithm, after step S30 is executed, the mathematical model based on the auction algorithm reaches a balance point, at this time, the allocation result is not the optimal allocation, therefore, the application proposes to repeatedly execute the bidding matching through the cyclic matching mode, so that the allocation result is constantly approaching the optimal allocation.

[0070] The application controls the cyclic iteration through the increment parameter ε, and dynamically attenuates the increment parameter ε in each cyclic matching , and terminates the cycle when , so as to ensure that the final matching result is basically consistent with the optimal matching result.

[0071] S50, the increment parameter ε is updated as ε=k, the current cost of each target point is reserved, the set of drones to be allocated U and the set A of allocation are reset;

[0072] (1) the increment parameter ε is updated as ε=k;

[0073] (2) the current cost of each target point is reserved, that is, the updated target point cost set P is reserved;

[0074] (3) the set of drones to be allocated U is reset, and all drones are re-allocated to the set of drones to be allocated U;

[0075] (4) the set A of allocation is reset, so that .

[0076] S60, the mathematical model based on the auction algorithm is solved by using the auction algorithm to perform the bidding matching on each drone, the set of drones to be allocated U, the target point cost set P and the set A of allocation are updated based on the matching result, and it is jumped to step S40.

[0077] After the increment parameter ε is updated and the set of drones to be allocated U and the set A of allocation are reset, the mathematical model based on the auction algorithm is solved again by using the auction algorithm to perform the bidding matching on each drone.

[0078] The specific process of performing the bidding matching on each drone in this step is the same as that of performing the bidding matching on each drone in step S30.

[0079] That is, in the step S60, each UAV is matched for bidding, and the specific process includes:

[0080] S61, selecting P UAVs from the set U of UAVs to be allocated;

[0081] S62, using an auction algorithm to perform parallel bidding matching on the selected P UAVs, and updating the set U of UAVs to be allocated, the set P of target point costs, and the set A of allocations based on the matching results.

[0082] In the step S62, the selected P UAVs perform operations in parallel:

[0083] S621, finding the target point j with the highest benefit for the UAV i, and calculating the highest benefit:

[0084] ,

[0085] wherein, represents the task benefit of the target point j for the UAV i, represents the current cost of the target point j.

[0086] S622, finding the target point m with the second highest benefit for the UAV i, and calculating the second highest benefit:

[0087] ,

[0088] wherein, represents the task benefit of the target point m for the UAV i, represents the current cost of the target point m.

[0089] S623, the UAV i bids for the target point j:

[0090] ,

[0091] wherein, represents the current cost of the target point j, is the highest benefit found, is the second highest benefit found, is a set incremental parameter.

[0092] S624, allocating the target point j to the UAV i, and storing the allocation in the set A of allocations, and removing the UAV from the set U of UAVs to be allocated.

[0093] In the step S624, when the target point j is allocated to the UAV i, if the target point j has been allocated to other UAVs, the previous allocation is removed, and the UAV previously allocated by the target point j is moved from the set A of allocations to the set U of UAVs to be allocated.

[0094] S625, updating the generation value of the target point j:

[0095] ,

[0096] wherein, is the bidding price of the UAV i for the target point j.

[0097] The step S62 is to perform the bidding matching for the P UAVs simultaneously, and for the steps S621 to S625, the target point j is locked by means of a global lock during the steps S623 of bidding by the UAV i for the target point j and S624 of assigning the target point j to the UAV i, so as to ensure that no data conflict occurs during parallel execution.

[0098] After the P UAVs to be selected complete the bidding matching by means of the steps S62 (including the steps S621 to S625), it is judged whether the set U of UAVs to be assigned is an empty set, if yes, the iteration is ended, and if not, the iteration is continued by jumping to the step S61, i.e., the steps S61 to S62 are repeatedly executed until the set U of UAVs to be assigned is an empty set.

[0099] In the step S60, after the bidding matching for each UAV is completed, the step S40 is jumped to, and it is judged whether the condition for ending the loop is reached.

[0100] The fast target assignment method suitable for large-scale UAV groups provided by the application can realize the following beneficial technical effects compared with the conventional target assignment algorithm:

[0101] (1) By introducing the auction algorithm, the parallel bidding mechanism of the auction algorithm is utilized, so that multiple UAVs independently perform target search and bidding decision each time, the parallel characteristic of the auction algorithm is fully utilized, the matching convergence speed is greatly accelerated, the matching time is greatly reduced, and the fast target matching of large-scale UAV groups is realized.

[0102] (2) By means of the cyclic matching mode and the termination of the cycle when the increment parameter is dynamically attenuated each time the cyclic matching is performed, the solving speed is accelerated, the assignment quality is improved, and the final matching result is basically consistent with the optimal matching result;

[0103] (3) Unlike the conventional target assignment algorithm which only uses the distance norm as the task benefit, the application constructs a new task benefit:

[0104] .

[0105] The advantages of the construction mode are:

[0106] ​(3.1) By the way of distance norm square, ensure that according to the matching result, the minimum distance between the straight line track of each UAV to its target point There is a lower bound, which reduces the pressure for subsequent path planning, that is,

[0107] ,

[0108] Wherein, represents the minimum distance between any two UAVs at the starting position, represents the minimum distance between any two UAVs at the target position;

[0109] (3.2) Multiply by on the basis of distance norm square , combined with the cyclic matching of , so that the final assignment result can basically keep consistent with the optimal matching result.

[0110] In order to test the performance of the application of the fast target assignment method for large-scale UAV group, four different scales of 200, 1000, 7000 and 11800 UAV groups are tested respectively (of which 7000 contains two different test cases, represented by 7000-A and 7000-B respectively). In the test, the initial position and target position of each flight are randomly disturbed, and the Hungarian algorithm and the fast target assignment method of the application are executed 1000 times respectively, and the average value of the total test time is taken, and the test results are as follows:

[0111] It should be noted that from the test results in the table, when the test flight is 200 times, the processing time of the Hungarian algorithm is shorter than that of the fast target assignment method of the application, because the fast target assignment method of the application needs to be matched in cycles, which leads to the occupation of additional time when processing the matching of UAV group with small flight such as 200 times. But when processing the matching problem of UAV group with more than 1000 times, the processing time of the fast target assignment method of the application is obviously accelerated.

[0112] Figure 2 Further shows the comparison test results of 11800 times.

[0113] Before describing the comparison test results of Figure 2 , define the variables distance square and M:

[0114] ;

[0115] Wherein, represents the position information of UAV i, represents the position information of the matching target point j obtained based on the target assignment algorithm, and n is the total number of UAVs.

[0116] The red line in the figure is the distance square sum (M1) based on the assignment result when the Hungarian algorithm matching is terminated, and the blue point is the distance square sum (M2) based on the assignment result of the fast target matching method. Figure 2 The ratio of the contrast test results of 11800 times is shown, and the ratio point in the figure is the ratio of M2 and M1, that is, M2 / M1. Figure 3 The relative error of the contrast test results of 11800 times is shown, and the error point in the figure is the ratio of (M2-M1) / M1. When M2 / M1 is closer to 1, (M2-M1) / M1 is closer to 0, which indicates that the fast target matching method is closer to the optimal solution (the matching result of the Hungarian algorithm is the optimal solution). From the figure, it can be seen that the fast target matching method is closer to the optimal solution. Figure 4 The relative error of the contrast test results of 11800 times is shown, and the error point in the figure is the ratio of (M2-M1) / M1. When M2 / M1 is closer to 1, (M2-M1) / M1 is closer to 0, which indicates that the fast target matching method is closer to the optimal solution (the matching result of the Hungarian algorithm is the optimal solution). From the figure, it can be seen that the fast target matching method is closer to the optimal solution. Figure 3 The relative error of the contrast test results of 11800 times is shown, and the error point in the figure is the ratio of (M2-M1) / M1. When M2 / M1 is closer to 1, (M2-M1) / M1 is closer to 0, which indicates that the fast target matching method is closer to the optimal solution (the matching result of the Hungarian algorithm is the optimal solution). From the figure, it can be seen that the fast target matching method is closer to the optimal solution. Figure 4 It can be seen that the fast target matching method and the result obtained by the Hungarian algorithm matching are basically consistent.

[0117] From the test results, compared with the Hungarian algorithm, the fast target assignment method proposed in the application greatly improves the solving speed when processing large-scale unmanned aerial vehicle group target assignment tasks, and ensures that the assignment result is basically consistent with the optimal matching result (the relative error of the final matching result is almost 0, and the average ratio is almost 1), which greatly improves the matching efficiency in the large-scale unmanned aerial vehicle performance path planning process.

[0118] The application further provides a fast target assignment system suitable for a large-scale unmanned aerial vehicle group, comprising at least one processor and at least one memory in communication connection with the processor, wherein the memory stores program instructions, and the program instructions are called and executed by the processor to realize Figure 1 The steps of the fast target assignment method suitable for a large-scale unmanned aerial vehicle group in the embodiment shown.

[0119] The application further provides a computer readable storage medium, which internally stores program instructions, and the program instructions are executed by a processor to realize Figure 1 The steps of the fast target assignment method suitable for a large-scale unmanned aerial vehicle group in the embodiment shown.

[0120] The above only describes the preferred embodiments of the application, and does not make any form of limitation on the application. Those skilled in the art can make various equivalent changes and improvements on the basis of the above embodiments, and any equivalent changes or modifications made within the scope of the claims shall fall within the protection scope of the application.

Claims

1. A fast target assignment method suitable for large-scale UAV swarm, characterized in that, The method comprises the following steps: S10, constructing a mathematical model based on an auction algorithm for a UAV group target matching problem; S20, initializing a UAV set to be allocated U, a target point cost set P, an allocation set A, an increment parameter ε, an increment scaling ratio k, and a task benefit of a target point; S30, solving the mathematical model based on the auction algorithm to perform bidding matching for each UAV, and updating the UAV set to be allocated U, the target point cost set P, and the allocation set A based on a matching result; S40, determining whether the increment parameter ε is greater than 1, if yes, performing step S50, and if no, returning to the allocation set A; S50, updating the increment parameter ε = ε / k, retaining a current cost of each target point, resetting the UAV set to be allocated U and the allocation set A; S60, solving the mathematical model based on the auction algorithm to perform bidding matching for each UAV, updating the UAV set to be allocated U, the target point cost set P, and the allocation set A based on a matching result, and jumping to step S40; In step S10, the mathematical model based on the auction algorithm is expressed as: ； Where P is the set of target point costs, including the current cost of all target points; Indicates the target point For drones Task rewards; Indicates the target point The current cost, where n is the number of drones; The step S30 further comprises: S31, selecting P UAVs from the UAV set to be allocated U; S32, performing bidding matching in parallel on the selected P UAVs by using the auction algorithm, and updating the UAV set to be allocated U, the target point cost set P, and the allocation set A based on a matching result; Steps S31 to S32 are repeatedly performed until the UAV set to be allocated U is empty; The step S32 further comprises: S321, finding a target point j with the highest benefit for a UAV i, and calculating the highest benefit: ； Wherein, n is the number of UAVs; S322, finding a target point m with the second highest benefit for the UAV i, and calculating the second highest benefit: ； wherein, represents the task benefit of the target point m to the UAV i, represents the current cost of the target point m; S323, the UAV i bidding for the target point j: ； S324, allocating the target point j to the UAV i, and storing the allocation in the allocation set A, and removing the UAV from the UAV set to be allocated U; S325, updating the cost value of the target point j: ； wherein, is the bid price of the drone i for the target point j.

2. The fast target assignment method suitable for large-scale drone swarm of claim 1, wherein, In the step S324, when the target point j is allocated to the UAV i, if the target point j has been allocated to other UAVs, the UAV previously allocated by the target point j is removed from the allocation set A to the UAV set to be allocated U.

3. The fast target assignment method suitable for large-scale drone swarm of claim 1, wherein, The step S20 further comprises: S21, obtaining UAV information, including id information of the UAV and position information of the UAV; S22, constructing and initializing the UAV set to be allocated U, the initialized UAV set to be allocated U including id information of all UAVs; S23, constructing and initializing the target point cost set P, the cost value of all target points in the initialized target point cost set P being 0; S24, constructing and initializing the allocation set A, the initialized allocation set A being an empty set; S25, setting the increment parameter ε; S26, setting the increment scaling ratio k, k > 1; S27, initializing the task benefit of each target point.

4. The fast target assignment method suitable for large-scale drone swarm of claim 3, wherein, The step S27 calculates the task benefit of each target point by using the following formula: ； wherein, representing a target point to the unmanned aerial vehicle task benefits, representing position information of the unmanned aerial vehicle representing position information of the target point representing a target point n is the number of unmanned aerial vehicles.

5. The fast target assignment method suitable for large-scale drone swarm of claim 3, wherein, The increment parameter initially set in the step S25 is ε > n / 10, n being the number of UAVs.

6. A fast target assignment system suitable for large-scale swarms of UAVs, characterized in that, The application discloses a computer readable storage medium, which comprises at least one processor and at least one memory connected with the processor in communication, wherein the memory stores program instructions, and the program instructions are called and executed by the processor to realize the method for quickly assigning targets to a large-scale unmanned aerial vehicle group according to any one of claims 1 to 5.

7. A computer readable storage medium characterized in that, The computer readable storage medium internally stores program instructions, and the program instructions are executed by the processor to realize the method for quickly assigning targets to a large-scale unmanned aerial vehicle group according to any one of claims 1 to 5.

Images