阅读说明：本技术 一种无人机群通联与覆盖组合优化方法 (Unmanned aerial vehicle group communication and coverage combination optimization method ) 是由 柴兴华 耿虎军 陈勇 于 2021-07-30 设计创作，主要内容包括：本发明公开了一种无人机群通联与覆盖组合优化方法,属于无人机任务规划技术领域。本发明首先确定无人机群通信保障任务的先验信息,然后采用蜂窝正六边形策略对任务区域进行分割,接着形成对全部目标的优化覆盖策略,并形成对机间通联关系的保持策略,最后完成无人机的最优指派。本发明在无人机通信范围有限的情况下,通过最优的无人机指派策略,实现数量及分布不均的多个目标的全覆盖,并能保证无人机之间的相互通联。(The invention discloses an unmanned aerial vehicle group communication and coverage combination optimization method, and belongs to the technical field of unmanned aerial vehicle mission planning. The method comprises the steps of firstly determining prior information of a communication guarantee task of the unmanned aerial vehicle cluster, then segmenting a task area by adopting a cellular regular hexagon strategy, then forming an optimized coverage strategy for all targets, forming a maintenance strategy for communication relation between the unmanned aerial vehicles, and finally finishing optimal assignment of the unmanned aerial vehicle. Under the condition that the communication range of the unmanned aerial vehicles is limited, the invention realizes the full coverage of a plurality of targets with uneven quantity and distribution by an optimal unmanned aerial vehicle assignment strategy, and can ensure the mutual communication among the unmanned aerial vehicles.)

1. A method for optimizing the communication and coverage combination of an unmanned aerial vehicle group is characterized by comprising the following steps:

step 1: determining prior information of a communication guarantee task of the unmanned aerial vehicle group, wherein the prior information comprises a task area, the current position of the unmanned aerial vehicle group, the communication range of the unmanned aerial vehicle and the ground target position;

step 2: determining the size parameters of a cellular regular hexagon according to the communication range of the unmanned aerial vehicle, and segmenting the task area determined in the step 1 by adopting the cellular regular hexagon strategy;

and step 3: optimizing the task area segmentation offset by taking the ground target which is most covered by the least honeycomb regular hexagon as an optimization index, and calculating the minimum number of unmanned aerial vehicles and expected positions required by covering all targets to form an optimized covering strategy for all targets;

and 4, step 4: counting the disconnected state between the unmanned aerial vehicles according to the expected positions of the unmanned aerial vehicles, and further taking the minimum number of the unmanned aerial vehicles as an optimization index according to the counting result to generate the number and the expected positions of the relay unmanned aerial vehicles so as to form a maintenance strategy for the communication relation between the unmanned aerial vehicles;

and 5: and according to the current position of the unmanned aerial vehicle cluster to be assigned and the expected position of the unmanned aerial vehicle cluster, taking the global shortest path as an optimization index, constructing the optimal assignment relation of each unmanned aerial vehicle, and finishing assignment.

2. The unmanned aerial vehicle fleet communication and coverage combined optimization method according to claim 1, wherein the task area is a convex polygon, and the specific manner of the step 2 is as follows:

step 2.1: sequentially taking each side of the convex polygon as a bottom side, and making parallel lines of the bottom side, wherein the parallel lines and the bottom side clamp the convex polygon together; calculating the distance between each bottom edge and the parallel line thereof, and taking the minimum value of the distances as the minimum height of the convex polygon;

step 2.2: setting the side length a of the honeycomb regular hexagon as the communication coverage radius R of the unmanned aerial vehicle_dI.e. the distance of the drone to the center of the cellular regular hexagon is R_d；

Step 2.3: establishing a rectangular coordinate system O-XY by taking the bottom edge corresponding to the minimum height of the convex polygon as an X axis and the left end point of the bottom edge as the origin of coordinates, and performing honeycomb segmentation on the whole task area under the coordinate system; wherein the coordinate origin coincides with the leftmost vertex of a honeycomb regular hexagon, the X axis is parallel to one side of the honeycomb regular hexagon, and the side length of the honeycomb regular hexagon is R determined in step 2.2_d。

3. The method as claimed in claim 2, wherein the specific manner of step 3 is as follows:

step 3.1: in the coordinate system determined in step 2.3, assuming that the number of drones required to cover all targets after Δ X and Δ Y movements of the cellular segmentation mesh in the X and Y directions is minimal,

step 3.2: and performing iterative optimization calculation on the delta x and the delta y by adopting the following formula:

wherein the content of the first and second substances,for iterative optimization times, omega is an inertia factor and takes a value of 0.4-0.9, T is the target number to be covered by communication, and delta x_t、Δy_tRespectively representing the two-dimensional offset of the t-th iteration,is a constant number, N_tRepresenting a two-dimensional offset of deltax_t、Δy_tThe number of the unmanned aerial vehicles required in the time is more than or equal to 1 and less than or equal to N_t≤T，r_1t、r_2tIs [0,1 ]]Random number in between, min (N)_0-t) And max (N)_0-t) Respectively represent N₀～N_tMinimum and maximum values of;

step 3.3: when the iterative optimization calculation of the step 3.2 meets the shutdown conditionOr a maximum number of iterations t is reached_stopTerminating the iteration in time, wherein lambda is a constant; outputting a final offset calculation result;

step 3.4: translating the cellular segmentation grids according to the final offset calculation result obtained in the step 3.3, wherein the number of cellular regular hexagons with targets after translation is the number N of unmanned aerial vehicles required for covering all targets, and the centers of the cellular regular hexagons are expected positions required for assigning communication coverage unmanned aerial vehicles;

step 3.5: and (4) generating an unmanned aerial vehicle expected position sequence according to the number and the expected positions of the unmanned aerial vehicles obtained in the step (3.4) to obtain an optimized coverage strategy for all targets.

4. The unmanned aerial vehicle fleet connectivity and coverage combined optimization method according to claim 3, wherein the specific manner of the step 4 is as follows:

step 4.1: according to the communication state among the unmanned aerial vehicles, dividing the unmanned aerial vehicle group into M disconnected sub-networks, and assigning relay unmanned aerial vehicles among the sub-networks for communication relay;

step 4.2: solving paths needing to relay the unmanned aerial vehicle fewest in all feasible communication paths to obtain a multi-region shortest path communication problem, and solving by adopting a Prim minimum spanning tree algorithm to obtain M-1 relay links for realizing communication;

step 4.3: calculating the number n of relay unmanned aerial vehicles required by the ith relay link through the following formula_i：

Wherein the content of the first and second substances,represents rounding up, d_iThe length of the ith relay link;

the total amount n of relay drones is:

n is carried out on the ith relay link_iAnd +1, equally dividing, namely the expected position of the relay unmanned aerial vehicle on the ith relay link at the equally dividing point.

5. The unmanned aerial vehicle fleet connectivity and coverage combined optimization method according to claim 4, wherein the specific manner of the step 5 is as follows:

step 5.1: according to the expected positions of all the unmanned aerial vehicles obtained in the steps 3 and 4 and the current position of the unmanned aerial vehicle cluster, defining the unmanned aerial vehicle assignment relationship between the expected positions and the current position as phi ═ phi₁,φ₂,...,φ_N+n]Wherein N + N is the number of all required unmanned aerial vehicles;

step 5.2: and taking the global shortest path as an optimization index, performing task allocation on the unmanned aerial vehicle based on a combined optimization method, solving phi, and finishing assignment according to a solving result.

Technical Field

The invention relates to an unmanned aerial vehicle group communication and coverage combined optimization method, belongs to the technical field of unmanned aerial vehicle task planning, and can be used for an unmanned aerial vehicle group communication target coverage task.

Background

With the continuous development of unmanned system technology, the unmanned aerial vehicle is widely applied to various fields of military and civilian with the advantages of low construction cost, high safety factor, unmanned operation and the like. However, because the single unmanned aerial vehicle has problems of long period, single task, poor survivability and the like in executing tasks, it is increasingly difficult to meet the task requirements of emergency rescue, target search, saturation attack and the like, and it is a necessary trend that the unmanned aerial vehicle group cooperatively completes various tasks. The method is a typical application, namely, the unmanned aerial vehicle cluster is adopted to construct an aerial communication network for targets (personnel, equipment, radio stations and the like) which are discretely distributed on the ground, so that the ground targets can be interconnected and communicated. In the task process, factors such as the communication coverage of the unmanned aerial vehicle cluster, the number of unmanned aerial vehicles, the group connectivity of the unmanned aerial vehicles, the number of covered targets and the like need to be considered, and the purpose of adopting fewer unmanned aerial vehicles to achieve higher coverage rate is achieved.

In the communication guarantee task of the unmanned aerial vehicle cluster, an important link is how to assign the minimum unmanned aerial vehicles to proper positions, realize the communication coverage of target points, and simultaneously keep the communication state among the unmanned aerial vehicle clusters. Current research is mainly focused on target coverage, considering the less connectivity constraint of unmanned clusters, and usually adopts target coverage methods such as K-means clustering ("K-means clustering algorithm for optimizing initial clustering center", 2020, computer engineering and application, 56 (15): 172-), DBSCAN density clustering ("DBSCAN algorithm research based on grid and density ratio", 2018, Liaoning university, Master's academic thesis), mean shift clustering and Mini Batch K-means clustering ("application of Mini Batch K-means algorithm in remote sensing image classification", 2017, Ludong university proceedings, 33 (4): 359- "363). The K-means clustering algorithm does not consider the coverage shape and range, and meanwhile, the random selection of the clustering center can cause low communication coverage rate and poor stability of the unmanned aerial vehicle cluster; the number of classes unspecified by the DBSCAN is limited, and the number of unmanned aerial vehicles in the communication guarantee is limited, so that the coverage rate of a target point becomes low due to the number of classes specified in the algorithm; the mean shift clustering also does not specify the number of classes, and the coverage rate of a target point in the communication guarantee of the unmanned aerial vehicle cluster is still unsatisfactory. The Mini Batch K-means algorithm is a variant of the K-means algorithm, has higher convergence rate than the K mean algorithm, but has poor clustering effect, and obviously is not suitable for the communication guarantee of the unmanned aerial vehicle cluster.

In summary, the existing target coverage method is applied to the communication guarantee task of the unmanned aerial vehicle cluster, and the problems of low coverage rate on the target point, poor stability, no consideration of the connectivity of the unmanned aerial vehicle cluster, and the like generally exist.

Disclosure of Invention

In view of the above, the present invention provides a method for optimizing a communication and coverage combination of a drone swarm, which is oriented to a ground user target communication coverage scenario with non-uniform distribution, and can realize full coverage of a plurality of ground targets with non-uniform distribution by using the minimum number of drones through an optimal drone assignment strategy under the condition that the communication coverage range of each drone is limited, and simultaneously satisfy the constraint conditions that the ground target coverage range of each drone and the maximum communication distance between drones are limited, and the like.

In order to achieve the purpose, the invention adopts the technical scheme that:

a method for optimizing the communication and coverage combination of an unmanned aerial vehicle cluster comprises the following steps:

step 1: determining prior information of a communication guarantee task of the unmanned aerial vehicle group, wherein the prior information comprises a task area, the current position of the unmanned aerial vehicle group, the communication range of the unmanned aerial vehicle and the ground target position;

step 2: determining the size parameters of a cellular regular hexagon according to the communication range of the unmanned aerial vehicle, and segmenting the task area determined in the step 1 by adopting the cellular regular hexagon strategy;

and step 3: optimizing the task area segmentation offset by taking the ground target which is most covered by the least honeycomb regular hexagon as an optimization index, and calculating the minimum number of unmanned aerial vehicles and expected positions required by covering all targets to form an optimized covering strategy for all targets;

and 4, step 4: counting the disconnected state between the unmanned aerial vehicles according to the expected positions of the unmanned aerial vehicles, and further taking the minimum number of the unmanned aerial vehicles as an optimization index according to the counting result to generate the number and the expected positions of the relay unmanned aerial vehicles so as to form a maintenance strategy for the communication relation between the unmanned aerial vehicles;

and 5: and according to the current position of the unmanned aerial vehicle cluster to be assigned and the expected position of the unmanned aerial vehicle cluster, taking the global shortest path as an optimization index, constructing the optimal assignment relation of each unmanned aerial vehicle, and finishing assignment.

Further, the task area is a convex polygon, and the specific manner of step 2 is as follows:

step 2.1: sequentially taking each side of the convex polygon as a bottom side, and making parallel lines of the bottom side, wherein the parallel lines and the bottom side clamp the convex polygon together; calculating the distance between each bottom edge and the parallel line thereof, and taking the minimum value of the distances as the minimum height of the convex polygon;

step 2.2: setting the side length a of the honeycomb regular hexagon as the communication coverage radius R of the unmanned aerial vehicle_dI.e. the distance of the drone to the center of the cellular regular hexagon is R_d；

Step 2.3: establishing a rectangular coordinate system O-XY by taking the bottom edge corresponding to the minimum height of the convex polygon as an X axis and the left end point of the bottom edge as the origin of coordinates, and performing honeycomb segmentation on the whole task area under the coordinate system; wherein the coordinate origin coincides with the leftmost vertex of a honeycomb regular hexagon, the X axis is parallel to one side of the honeycomb regular hexagon, and the side length of the honeycomb regular hexagon is R determined in step 2.2_d。

Further, the specific manner of step 3 is as follows:

step 3.1: in the coordinate system determined in step 2.3, assuming that the number of drones required to cover all targets after Δ X and Δ Y movements of the cellular segmentation mesh in the X and Y directions is minimal,

step 3.2: and performing iterative optimization calculation on the delta x and the delta y by adopting the following formula:

wherein the content of the first and second substances,is an overlapReplacing optimization times, wherein omega is an inertia factor and takes a value of 0.4-0.9, T is the target number to be covered by communication, and delta x_t、Δy_tRespectively representing the two-dimensional offset of the t-th iteration,is a constant number, N_tRepresenting a two-dimensional offset of deltax_t、Δy_tThe number of the unmanned aerial vehicles required in the time is more than or equal to 1 and less than or equal to N_t≤T，r_1t、r_2tIs [0,1 ]]Random number in between, min (N)_0-t) And max (N)_0-t) Respectively represent N₀～N_tMinimum and maximum values of;

step 3.3: when the iterative optimization calculation of the step 3.2 meets the shutdown conditionOr a maximum number of iterations t is reached_stopTerminating the iteration in time, wherein lambda is a constant; outputting a final offset calculation result;

step 3.4: translating the cellular segmentation grids according to the final offset calculation result obtained in the step 3.3, wherein the number of cellular regular hexagons with targets after translation is the number N of unmanned aerial vehicles required for covering all targets, and the centers of the cellular regular hexagons are expected positions required for assigning communication coverage unmanned aerial vehicles;

step 3.5: and (4) generating an unmanned aerial vehicle expected position sequence according to the number and the expected positions of the unmanned aerial vehicles obtained in the step (3.4) to obtain an optimized coverage strategy for all targets.

Further, the specific manner of step 4 is as follows:

step 4.1: according to the communication state among the unmanned aerial vehicles, dividing the unmanned aerial vehicle group into M disconnected sub-networks, and assigning relay unmanned aerial vehicles among the sub-networks for communication relay;

step 4.2: solving paths needing to relay the unmanned aerial vehicle fewest in all feasible communication paths to obtain a multi-region shortest path communication problem, and solving by adopting a Prim minimum spanning tree algorithm to obtain M-1 relay links for realizing communication;

step 4.3: calculating the number n of relay unmanned aerial vehicles required by the ith relay link through the following formula_i：

Wherein the content of the first and second substances,represents rounding up, d_iThe length of the ith relay link;

the total amount n of relay drones is:

n is carried out on the ith relay link_iAnd +1, equally dividing, namely the expected position of the relay unmanned aerial vehicle on the ith relay link at the equally dividing point.

Further, the specific manner of step 5 is as follows:

step 5.1: according to the expected positions of all the unmanned aerial vehicles obtained in the steps 3 and 4 and the current position of the unmanned aerial vehicle cluster, defining the unmanned aerial vehicle assignment relationship between the expected positions and the current position as phi ═ phi₁,φ₂,...,φ_N+n]Wherein N + N is the number of all required unmanned aerial vehicles;

step 5.2: and taking the global shortest path as an optimization index, performing task allocation on the unmanned aerial vehicle based on a combined optimization method, solving phi, and finishing assignment according to a solving result.

The invention has the beneficial effects that:

1. the invention solves the minimum unmanned aerial vehicle quantity and the corresponding position required by the target coverage by adopting a shortest path combination optimization method among a plurality of local areas through strategies of optimizing the internal relation between the coverage radius of the unmanned aerial vehicle and the map segmentation, analyzing the influence of the map segmentation offset on the target optimization coverage, keeping the constraint of the connectivity of the unmanned aerial vehicle cluster and the like, and realizes the optimal assignment of the unmanned aerial vehicle cluster by taking the shortest flight path of the group as an optimization index.

2. The invention fully utilizes the discrete characteristic in the process of ensuring the task execution of the unmanned aerial vehicle cluster communication, converts the coverage problem of a large-scale non-uniform dynamic target into a multi-target combination optimization problem, converts the unmanned aerial vehicle cluster communication maintenance strategy adjustment into the problem of solving the unmanned aerial vehicle node dynamic assignment strategy, and can realize the complex task allocation of multiple unmanned aerial vehicles through simple calculation, so that the method has stronger scene adaptability and is beneficial to engineering application.

Drawings

FIG. 1 is a flow chart of a method for optimizing communication and coverage combinations of an unmanned aerial vehicle fleet;

FIG. 2 is a schematic diagram of a method for determining a minimum height of a polygonal task area;

FIG. 3 is a schematic diagram of a size parameter calculation method of a honeycomb regular hexagon;

FIG. 4 is a schematic diagram of task area cell segmentation;

FIG. 5 is a diagram illustrating the effect of cell offset on the optimization solution, wherein (a) is a diagram illustrating initial cell hexagon partition, and (b) is a diagram illustrating cell hexagon partition after translation;

FIG. 6 is a statistical schematic diagram of a disconnected sub-network of the unmanned aerial vehicle cluster;

FIG. 7 is a graph of simulation results;

FIG. 8 is a schematic illustration of a simulation experiment.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings.

An optimization method for the communication and coverage combination of an unmanned aerial vehicle cluster is applied to an unmanned aerial vehicle cluster system for communication guarantee tasks, and the system can be described as follows: aiming at the communication guarantee task of the ground users, an unmanned aerial communication channel is established for the ground users in a self-organizing mode by taking an unmanned aerial vehicle cluster formed by a plurality of isomorphic unmanned aerial vehicles as a carrier, so that the ground users which are non-uniformly distributed can be interconnected and communicated through the aerial channel, and the premise of realizing the communication guarantee task is that the target position of the ground user can be acquired by an unmanned aerial vehicle cluster system in real time.

As shown in fig. 1, the method comprises the steps of:

step 1: and determining prior information of the communication guarantee task of the unmanned aerial vehicle group. The method mainly comprises a task area, the current position of an unmanned aerial vehicle cluster, the communication range of the unmanned aerial vehicle, the ground target position and the like.

Wherein, 1) the task area is usually represented by a convex polygon, which is intended to define a range for the target to be covered by communication, and the target only appears in the area or the target appearing outside the area is not in the execution range of the communication coverage task; 2) the unmanned aerial vehicle communication range comprises a communication relay range (the longest distance kept by communication between the unmanned aerial vehicles and represented by a circle with the unmanned aerial vehicle as the center) and a communication coverage range (the longest distance covering ground target communication and represented by a projection circle of the circle with the unmanned aerial vehicle as the center on the ground), is usually determined by unmanned aerial vehicle communication load and is constant under the condition of communication load determination; 3) the ground target position has randomness and motion continuity, and can be acquired by the unmanned aerial vehicle group system in real time through a positioning means, and the target position is continuously acquired in the whole task implementation process and is used as the input of the designated unmanned aerial vehicle.

Step 2: and (3) determining the size parameters of the cellular regular hexagon according to the communication range of the unmanned aerial vehicle, and segmenting the task area determined in the step (1) by adopting the cellular regular hexagon strategy.

2.1 determining the origin of the task area according to the minimum height of the polygonal task area. As shown in FIG. 2, for the irregular convex polygon task execution area, assuming convex pentagonal ABCDE, the optimal height of the polygon is measured by making parallel lines, one line of the parallel lines is sequentially overlapped with one side of the polygon, the optimal height can be measured step by step according to DC, CB, BA, AE and ED, the other line crosses over the vertex of the opposite side, and the minimum span of all the parallel lines is called as the minimum height H of the convex polygonal ABCDE_minThe vertex is D and the edge is BA. The origin of the task area is the lower left vertex B of the polygon area in this state.

2.2 calculating the size parameters of the honeycomb partition regular hexagon. Suppose the coverage radius of the drone is R_dTo achieve coverage of target points and minimize overlapping coverage, and to further simplify queriesSolving the problem, setting the side length a of the hexagon as R_dWhen assigning a drone to a hexagonal center location, coverage of targets within the area is achieved, as shown in fig. 3. Under this kind of division, unmanned aerial vehicle will cover complete hexagon honeycomb at every turn, and unmanned aerial vehicle shift position is the honeycomb center of discretization, and then has simplified subsequent optimization problem solution under the condition of guaranteeing higher coverage precision.

And 2.3, taking the original point of the task area determined in the step 2.1 as the coordinate original point, taking the bottom edge of the task area as an X axis, establishing a rectangular coordinate system O-XY, taking the side length a of the hexagon of the honeycomb calculated in the step 2.2 as a basic size in the coordinate system, overlapping the leftmost vertex of the hexagon with the original point O of the coordinate, and dividing the honeycomb into the whole task area, wherein the dividing effect is shown in FIG. 4.

And step 3: and optimizing the cellular segmentation offset by using the ground target which is covered most by the least cellular area as an optimization index, and calculating the minimum number of unmanned aerial vehicles and expected positions required by covering all targets.

3.1 setting up under the coordinate system determined in step 2.3, moving the cellular map by delta X and delta Y in X direction and Y direction, limiting to avoid periodical coincidence caused by unlimited offset

As shown in fig. 5, for the same distributed target positions, five unmanned aerial vehicles are needed to cover the target points in the graph (a) dividing mode, and after the offset of cellular division is properly adjusted, as shown in the graph (b), two target nodes originally distributed in different cells are adjusted to the same cellular region, only four unmanned aerial vehicles are needed to complete the coverage (the communication relation is not considered temporarily), so that the discrete offset of the hexagonal map cells is reasonably optimized, the resource allocation can be further optimized, and the usage amount of the unmanned aerial vehicles is reduced.

3.2 setting a mixed stopping criterion, judging whether to stop iterative computation according to the stopping criterion, wherein the mixed stopping criterion adopted by the scheme isWherein λ is a constant; setting a maximum number of iterations t at the same time_stopThe objective is to prevent the iterative process from misconvergence, where the fitness outage criterion has the highest priority, the largest number of iterations being the second.

3.3 based on the shutdown criteria described in step 3.2, the following equations are used to perform iterative optimization calculations for Δ x, Δ y:

wherein the content of the first and second substances,optimizing the number of times for iteration; omega is an inertia factor, and the value of omega is generally between 0.4 and 0.9; t is the target number to be covered by communication; Δ x_t、Δy_tRespectively representing t-time iteration two-dimensional offset;is a constant; n is a radical of_tRepresenting a two-dimensional offset of deltax_t、Δy_tThe number of the unmanned aerial vehicles required in the process is more than or equal to 1 and less than or equal to N_t≤T；r_1t、r_2tIs [0,1 ]]A random number in between; min (N)_0-t) And max (N)_0-t) Respectively represent N₀～N_tMinimum and maximum values of (d).

3.4 if the stop condition described in step 3.2 is met, i.e.Or t ═ t_stop-1, stopping the search and outputting the final offset calculation result Δ x ═ Δ x_t+1、Δy＝Δy_t+1Translating the cellular grid according to the result, wherein the cellular grid center with the target after translation is the expected position of the unmanned aerial vehicle needing to be assigned for communication coverage; otherwise, returning to step 3.3 to continue to iterate the optimization offset deltax_t、Δy_t。

3.5 according to the number of the needed unmanned aerial vehicles, taking the honeycomb grid center with the target to be covered as a target point, generating an unmanned aerial vehicle expected position sequence, and generating an optimized coverage strategy for all the targets.

And 4, step 4: according to the sequence of the expected positions of the unmanned aerial vehicles, counting the disconnected conditions among the unmanned aerial vehicles, and further taking the minimum number of the unmanned aerial vehicles as an optimization index according to the result, generating the expected positions specially used for the communication relay unmanned aerial vehicles, and forming a maintenance strategy for the communication relation among the unmanned aerial vehicles.

4.1 according to the communication state of the unmanned aerial vehicles, the unmanned aerial vehicle cluster is divided into M disconnected sub-networks, the network topology between the sub-networks is disconnected, and as shown in fig. 6, the unmanned aerial vehicles need to be assigned to be exclusively used for communication relay between the sub-networks.

4.2 solve the minimum number of routes of required communication UNICOM unmanned aerial vehicle in all feasible communication routes, turn into a multizone shortest path intercommunication problem with the problem. And solving the problem by adopting a Prim algorithm (minimum spanning tree), and obtaining M-1 paths for ensuring communication topology connection, wherein the sum of the paths is shortest, so that the number of unmanned aerial vehicles required to be supported is minimum, and the resource waste is further effectively reduced.

4.3 calculating the number n of communication relay unmanned aerial vehicles required by the ith relay link through the following formula_i：

Wherein the content of the first and second substances,represents rounding up, d_iIs the length of the ith relay link. The total number n of communication relay unmanned aerial vehicles needing to be assigned is as follows:

position passing pair d of communication relay unmanned aerial vehicle required to be assigned_iCarry out n_i+1 equal division determination.

And 5: according to the current position of the unmanned aerial vehicle cluster to be assigned and the expected position of the unmanned aerial vehicle cluster, the global shortest path is used as an optimization index, the optimal assignment relation of each unmanned aerial vehicle is constructed, and assignment is completed;

5.1 the target coverage or communication relay unmanned aerial vehicle expected position T ═ T output according to step 3.5 and step 5.3₁,t₂,...,t_N+n]Combining the current position X ═ X where the unmanned aerial vehicle group is located₁,x₂,...,x_N+n]Defining a drone assignment relationship between the desired location and the current location as phi ═ phi₁,φ₂,...,φ_N+n]Wherein N + N is the number of all required unmanned aerial vehicles;

and 5.2, taking the global shortest path as an optimization index, adopting a Hungarian algorithm to perform task allocation on the unmanned aerial vehicle, solving phi, and finishing assignment according to a solving result. The solution process of optimization by using the Hungarian algorithm can be seen in a Valley article, "study and application of target allocation problem based on evolutionary Hungarian algorithm" (university of electronic science and technology, Master academic thesis, 2013).

Step 6: when the target position and the distribution are not changed, the unmanned aerial vehicle group keeps the current state; and when the target position and the distribution are changed, repeating the steps 3-5 to execute a new round of assignment until the task is finished.

The effect of the method can be illustrated by the following simulation experiment:

(1) simulation implementation conditions

In the following simulation, the number of targets to be supported is 24, the task area is a convex quadrilateral, coordinates of four vertexes of the quadrilateral are (0,0), (140,90) and (0,90), namely, the value range of the abscissa of each target point is [0,140 ]]The longitudinal coordinate value interval is [0,90 ]]And the communication coverage radius of the unmanned aerial vehicle is R_dIs 14.14.

(2) Simulation implementation content and results

Randomly generating 24 target points to be covered, and calculating the number N of the target unmanned aerial vehicles to be covered to be 19 and the single-machine communication radius R according to the minimum required number and expected position calculation method of the communication coverage unmanned aerial vehicles provided by the patent_dAt a condition of 14.14, according to the method, the meterAnd calculating the number n of the communication relay unmanned aerial vehicles to be 6. The overall communication and coverage combined optimization result is shown in fig. 7, and the meaning of each label in the figure is shown in fig. 8.

Fig. 7 shows that, for a plurality of targets to be covered by communication, the method can achieve the purpose that the unmanned aerial vehicle cluster covers multiple target points to the maximum extent in the communication covering process under the condition that the positions of the target points are randomly generated, and meanwhile, the communication relation among multiple unmanned aerial vehicles is ensured, so that the method has strong universality and stable task effect.

In a word, the invention provides typical application of communication guarantee for the targets in a specific task area by aiming at the unmanned aerial vehicle group, and adopts the unmanned aerial vehicle group to construct an aerial communication network for the targets (personnel, equipment, radio stations, and the like) which are discretely distributed on the ground, so that the ground targets can be interconnected and intercommunicated. The invention fully considers the factors of unmanned aerial vehicle cluster communication coverage area, unmanned aerial vehicle number, unmanned aerial vehicle cluster connectivity, target coverage number and the like, and can realize higher coverage rate by adopting fewer unmanned aerial vehicles for the target.

The above description is only some specific examples of the present invention and should not be construed as limiting the invention in any way. It will be apparent to persons skilled in the relevant art that various modifications and changes in form and detail can be made therein without departing from the principles and arrangements of the invention, but such modifications and changes are within the spirit and scope of the invention as defined in the appended claims.