CN115801095A - Air-ground relay communication control method for unmanned aerial vehicle cluster application - Google Patents

Abstract

The invention discloses an air-to-ground relay communication control method oriented to the application of UAV clusters. The steps include: S01. Modeling the UAV communication process; S02. Obtaining the spatial distribution of task points and judging whether the current air-to-ground communication conditions require Relay, if necessary, aim at the least relay node, determine the number of relay UAVs and the spatial location of the relay UAV access; S03. Obtain the timing of the mission points in the cluster to maximize the UAV Construct the objective function for the relay communication income of the multi-mission point, construct the dynamic relay UAV mission decision model based on the objective function and the required constraints, and obtain the relay UAV mission decision model by solving the dynamic relay UAV mission decision model. The planning result of the task. The invention has the advantages of simple implementation method, low complexity, strong safety and stability, and can eliminate task point conflicts among multiple relay UAVs, and at the same time minimizes the number of relay UAVs required for cluster tasks of air-ground communication.

Description

Air-to-ground relay communication control method for UAV swarm applications

technical field

The invention relates to the technical field of unmanned aerial vehicle cluster communication control, in particular to an air-to-ground relay communication control method for unmanned aerial vehicle cluster applications.

Background technique

In the application process, the UAV cluster has the characteristics of dispersed UAV nodes and flying across mission areas. UAVs rely on data links to establish communications with ground control stations and other UAVs during mission execution, and transmit business data and mission information. , to maintain the cluster organizational structure and overall task execution. However, the distribution range of each mission point is wide and scattered, and there are problems of poor communication caused by long-distance (non-line-of-sight), terrain and buildings, etc., making the air-ground communication between the cluster and the ground control station impossible to guarantee . In this regard, the introduction of relay drones can solve the problem of blocked direct information transmission between air and ground.

Compared with communication based on ground base stations and satellite systems, as shown in Figure 1, UAV relay communication has obvious advantages: First, rapid deployment can be achieved. Establish short-range line-of-sight and medium-to-long-distance non-line-of-sight communication links to improve communication performance between communication nodes; secondly, the pose is easy to adjust, and the motion and attitude of the aerial UAV can be dynamically adjusted to adapt to changes in the communication environment (such as terrain shading, rainy days, etc.) Attenuation/communication noise or interference, etc.), which can improve the environmental adaptability of the communication link; finally, the maintenance cost is low, the operation and maintenance cost of the UAV system is lower, the take-off and recovery are more flexible, and it can be deployed at any time to start operations, suitable for Unexpected or short-duration tasks.

UAVs play an important role in assisting wireless communication, and there are three typical applications: (1) Regional coverage and emergency communication: Deploying UAVs to assist existing communication infrastructure to provide seamless wireless communication in service areas Coverage, its application scenarios include infrastructure damage due to natural disasters and base station overload in extremely congested areas; (2) Long-distance communication: UAV communication relay between two remote users or users without a reliable direct link Provide wireless connection between groups; (3) Information collection and distribution, UAV-assisted information distribution and data collection, dispatching UAVs to distribute to a large number of distributed wireless devices (for example, wireless sensors in farmland) or collect delays from wireless devices Tolerate the message. The UAV swarm has a wide range of mission areas. It is likely that the communication between the ground control station and the mission UAV will be blocked due to the distance between the ground control station and the mission UAV. With the help of the relay communication function of the relay drone, the command control message and the status information of the drone are forwarded.

However, due to its operating environment, movement mode and other characteristics, UAVs are affected by many factors such as terrain, weather, movement attitude and speed, and signal interference during the ground, sea and air communication relay tasks. Research on machine-to-machine relay communication is facing great challenges. In the prior art, when UAV mission planning and multi-machine online collaboration, usually only a distance constraint is set for UAV communication, and the channel model is not penetrated into the channel model. Changes will also cause changes in the communication topology of the inter-machine network, resulting in the pre- or online mission planning, collaborative control, and flight control that are not applicable to the changed inter-machine network communication topology during UAV cluster control. There may be conflicts between task points between machines.

For the relay communication planning in UAV cluster applications, it is difficult to eliminate the task point conflicts between multi-relay UAVs in the existing technology, and the number of relay UAVs required for cluster tasks is large, so it is urgent to provide An air-to-ground relay communication control method for unmanned aerial vehicle swarm applications, so as to eliminate mission point conflicts between multi-relay unmanned aerial vehicles in air-ground communication, and at the same time reduce the number of relay unmanned aerial vehicles for cluster tasks as much as possible.

Contents of the invention

The technical problem to be solved by the present invention lies in: aiming at the technical problems existing in the prior art, the present invention provides an air-to-ground relay communication control method oriented to UAV cluster applications with a simple implementation method, high efficiency and strong flexibility, which can Eliminate mission point conflicts between multi-relay UAVs, and realize the air-ground communication of the least-relay UAVs to ensure the cluster mission process.

In order to solve the problems of the technologies described above, the technical solution proposed by the present invention is:

An air-to-ground relay communication control method for unmanned aerial vehicle swarm applications, the steps comprising:

S01. Model the UAV communication, construct the path loss calculation model between the UAV and the ground control station, the communication model between the UAV and the ground station, and the communication model between the UAV and the relay UAV channel model;

S02. Static analysis of relay demand based on the spatial distribution of mission points: obtain the spatial distribution of mission points, judge whether the current air-to-ground communication conditions need to be relayed according to the obtained spatial distribution and the model constructed in step S01, and if necessary, use the minimum The relay node is used as the target to determine the number of relay UAVs and the spatial location of the relay UAV access;

S03. Planning the relay task based on the task sequence of the cluster UAV: obtain the sequence of the task points in the cluster, and construct an objective function with the goal of maximizing the relay communication revenue of the drone to the multi-task point, based on the objective function and the obtained The required constraint conditions are constructed to obtain a dynamic relay UAV mission decision model, and the planning result of the relay mission is obtained by solving the dynamic relay UAV mission decision model.

构建无人机和地面控制站的期望路径损耗Λ计算模型为：Further, in the step S1, the position of the ground control station is (x _bs , y _bs , h _bs ), the coordinates of the drone m are (x _m , y _m , h _m ), and the drone m is The elevation angle of the station is

Construct the expected path loss Λ calculation model of UAV and ground control station as:

η_LoS、η_NLoS分别为视距通信链路LoS、非视距通信链路NLoS的附加路径损耗，f_c为无线电波的载波频率，d_mb为无人机m和地面控制站的距离，

c为光波速度，α、β为环境参数；Among them, A=η _LoS -η _NLoS ,

_ηLoS and _ηNLoS are the additional path losses of the line-of-sight communication link LoS and the non-line-of-sight communication link NLoS respectively, f _c is the carrier frequency of radio waves, d _mb is the distance between the UAV m and the ground control station,

c is the speed of light wave, α, β are environmental parameters;

The total antenna gain calculation expression is:

Among them, κ is the total antenna gain, F(θ) is the antenna power gain, θ is the angle brought by the height difference of the transmitting and receiving antenna, and η is the angle formed by the transmitting and receiving antenna due to the change of the UAV's own attitude;

According to the signal path loss and antenna gain, the calculation expression of the total loss Loss of the air-to-ground channel signal is constructed as follows:

Loss=Λ-log 10(F(θ)·cosη).

Further, in the step S01, the communication model between the constructed UAV and the ground station is:

Wherein, A=η _LoS -η _NLOS , B=20log f _c +20log(4π/c)+η _NLOS , and SNR _th is the preset signal-to-noise ratio threshold;

The channel model between the unmanned aerial vehicle and the relay unmanned aerial vehicle is:

Further, in the step S02, the optimization goal is to deploy the least number of relay drones to meet the communication requirements between all mission points and the ground control station, and the decision variables are the number of relay drones and the corresponding relay deployment Position, by minimizing the number of relay drones as the static coverage optimization goal, the location of the relay drone and its association with the corresponding task area are determined, and the optimization problem formed is specifically:

Among them, C ₁ is the Boolean decision variable of the current problem, C ₂ indicates that the communication between the mission point and the ground control station can only be relayed by one UAV, and C ₃ indicates the number of mission points relayed by the relay UAV at the same time In the current upper bound, C ₄ restricts the mission point not to be associated with the unused relay UAV, C ₅ and C ₆ represent the distance requirements between the relay UAV and the associated mission point and the ground control station, L is When the preset value can make α _mk and u _m equal to 0, there is no restriction on the m distance of the relay UAV, x _re,m ,y _re,m represent the horizontal and vertical coordinates of the relay position, u=[u _m ] _1×K represents the state set of candidate relay UAVs, u _m = 1 means that the mth candidate relay UAV is adopted, u _m = 0 means delete the candidate UAV, R _bs Indicates the communication radius, α=[α _mk ] _K×K represents the correlation matrix between the relay UAV and the mission point, α _mk =1 means that the m-th relay UAV performs relay communication for the k-th mission point , that is, the mission point k is within the communication range of the relay UAV m, N _max represents the maximum number of mission UAVs connected to a relay UAV at the same time, x _ts,k ,y _ts,k represents the kth The horizontal and vertical coordinates of a task point.

Further, in the step S03, according to the deployment method of the static relay, UAV relays are arranged for all mission points at the same time, and the static deployment points of the UAV relays are obtained; the communication of the static relay deployment points is calculated. Time window, assign relay UAVs to traverse the relay static deployment points in turn according to the time window requirements, and integrate the income of the communication of the mission point itself, the path of the relay UAV, and the time for the relay UAV to reach the corresponding mission point. The time and energy consumption factors of the relay UAV are used to construct the objective function with the goal of maximizing the relay communication revenue of the UAV to the multi-mission point.

Further, the objective function constructed to maximize the relay communication revenue of the UAV to the multi-mission point is:

为中继无人机m和静态覆盖部署点的对应关系，β_mn＝1表示第n个静态覆盖部署点分配给了第m架中继无人机，L_m为分配给中继无人机m的静态覆盖部署点总数，

为第m架中继无人机的飞行路径，p_mk为中继无人机所属的静态覆盖部署点，

为对应p_m中每个静态覆盖部署点到达时间，τ_mk表示中继无人机m到达任务点p_mk的时间；Among them, the function c _mn (β _m ,p _m ,τ _m ) is the relay communication income obtained by the m-th relay drone arriving at the corresponding relay deployment point n according to the path p _m and mission time τ _m minus the flight path total energy consumption,

is the corresponding relationship between the relay UAV m and the static coverage deployment point, β _mn = 1 means that the nth static coverage deployment point is assigned to the mth relay UAV, and L _m is the distribution to the relay UAV The total number of static coverage deployment points for m,

is the flight path of the mth relay UAV, p _mk is the static coverage deployment point to which the relay UAV belongs,

To correspond to the arrival time of each static coverage deployment point in p _m , τ _mk represents the time when the relay UAV m reaches the mission point p _mk ;

The constraints include:

τ _mk ≤O _re,n1 ,τ _m(k+1) ≤O _re,n2 ,

(O _re,n2 -C _re,n1 )*v≥||(x _re,n1 -x _re,n2 ,y _re,n1 -y _re,n2 )|| ₂ ,

p _mk ＝n1,p _mk+1 ＝n2,

Among them, the starting point of the time window O _re,n is the earliest time of the starting point of all time windows in the corresponding task point set Ta _n , and the end point of the time window C _re,n is the latest time of the ending time of all task time windows in Ta _n , indicating the relay position The horizontal and vertical coordinates of .

Further, in the step S03, the consensus-based bundle set algorithm CBBA is used to solve the mission decision model of the dynamic relay UAV, and the two stages of task bundle construction and conflict elimination are continuously iteratively executed during the solution process until all The task distribution is completed, wherein in the stage of the task bundle construction, each relay drone creates a task bundle y to store the relay task points assigned by itself, the execution order of the relay task points and the task execution time, and Save the assignors corresponding to all tasks and the benefits that the tasks will bring, and update them continuously during the iterative process; in the phase of conflict elimination, the adjacent relay drones communicate with each other to compare the list of successful bidders stored separately and the bid-winning value list, updating the bid-winning value list according to the sequence of time stamps.

Further, in the phase of the task bundle construction, the relay UAV continuously adds relay task points to the task bundle until it reaches the limit of the maximum number of relay task points L _max of the relay UAV or there is no vacancy Relay points, each relay UAV maintains a list of task bundles, a list of path task points, and a list of arrival time of path task points; according to each list, calculate the increase in revenue that will be brought by relaying communication tasks at each insertion position amount, and take the position with the largest revenue increase as the path sequence corresponding to the new relay task point n, where the revenue increase brought by adding the relay task point n to the task bundle b _m of the relay UAV m is calculated according to the following formula quantity:

表示将新中继任务点n插入路径p_m上第δ个路径点前的总体收益，

表示按照路径p_m顺序执行中继任务将会带来的总收益；Among them, |p _m | is the number of task points assigned to the relay UAV,

Indicates the overall revenue of inserting the new relay task point n before the δth path point on the path p _m ,

Indicates the total revenue that will be brought by executing the relay tasks in sequence according to the path p _m ;

The expression for calculating the total revenue of the relay UAV m performing tasks along p _m is:

成正相关，第三个因素是一个惩罚项，以使得无人机m从上一个中继任务点b_m(k-1)到达当前任务点b_mk所需要消耗的燃料与两任务点间的距离成正比。Among them, q ₀ is the coordinates of the starting point of the relay UAV, v ₀ is the cruising speed of the relay UAV, s _mk (τ _mk ,b _mk ) is the arrival of the relay UAV m at the relay mission point at time τ _mk The income obtained when b _mk ; s _mk (τ _mk ,b _mk ) is determined by three factors, the first factor is the time to reach the mission point τ _mk , the second factor is the relay communication value Val _mk of the mission point itself , the sum of the communication time of the relay communication value Val _mk and all task points covered by the relay task point b _mk

is positively correlated, the third factor is a penalty item, so that the distance between the fuel consumed by the UAV m from the last relay mission point b _m(k-1) to the current mission point b _mk and the two mission points Proportional.

中继m₁与中继m₂最近一次通信时间则为消息接收时间t_r；当两者之间无直连通路时，查找中继无人机m₁直连的所有其余中继无人机

在查找到的中继无人机集合C中找到最近的时间戳

当中继无人机m₁从中继m₂处更新信息，将中继无人机m₁存储的中标者向量

和中标出价向量

融合更新。Further, in the stage of conflict elimination, the relay UAV adopts a consistency strategy to converge the bid-winning list of the relay task, and assigns the relay task point to be reached to the relay UAV according to the bid-winning list, thereby giving the relay task The mission UAV at the mission point corresponding to the point provides communication relay service; when there is a direct link between the relay UAV m ₁ and the relay m ₂ , that is

The latest communication time between relay m ₁ and relay m ₂ is the message receiving time t _r ; when there is no direct connection between the two, search for all other relay drones directly connected to relay drone m ₁

Find the most recent timestamp in the set of relay drones found C

When relay drone m ₁ updates information from relay m ₂ , the successful bidder vector stored in relay drone m ₁ will be

and the winning bid vector

Fusion update.

和中标者向量

当中继无人机m₁判定任务点n中标者是自身，且中继m₂判定中标者是中继i或者空，则不执行任何操作，中继m₁的中标向量保持不变，当两中继无人机存储的中标者产生冲突时，则对中继m₁两个中标向量进行重置。Further, in the stage of conflict elimination, if the winning bid of the relay task point n stored by the relay drone m ₂ is higher than that of the relay m ₁ , or one of the two relays stores a value for the relay task If the successful bidder of point n is m, m≠m ₁ ∩m≠m ₂ , and the timestamp of the latest message about relay m received by relay m ₂ is later than that of relay m ₁ , then relay drone m _1. Execute the update operation of the successful bid vector, and assign the successful bid vector and the successful bidder vector of relay m ₂ to the successful bid vector of relay m ₁

and the winning bidder vector

When the relay UAV m ₁ judges that the successful bidder of mission point n is itself, and the relay m ₂ judges that the successful bidder is relay i or empty, no operation is performed, and the winning vector of relay m ₁ remains unchanged. When the successful bidder stored in the relay UAV conflicts, the two successful bid vectors of the relay _m1 are reset.

Compared with the prior art, the present invention has the advantages that: the present invention considers the narrow-band real-time communication requirements between the UAV and the ground station, and the broadband high-speed communication requirements during cooperative operations between the aircraft, and first bases on the spatial distribution of mission points to minimize The relay node is the target, determine the number of relay drones that need to be relayed and the spatial location of the relay drone access, and then maximize the relay communication revenue of the drone to the multi-mission point according to the task sequence of the cluster drone The objective function is constructed for the target, and then the dynamic relay UAV task decision model is constructed. By solving the dynamic relay UAV mission decision model for relay planning, the relay UAV path planning and online Path adjustment and other requirements, realize the continuous and stable transmission of command information between the UAV and the ground and the coordination information between the UAVs, not only can eliminate the conflict of mission points between multi-relay UAVs, but also can make the air-ground communication cluster tasks required Relay drones have the fewest number.

Description of drawings

Figure 1 is a schematic diagram of the principle of the UAV relay communication architecture.

FIG. 2 is a schematic diagram of the key flow of the air-to-ground relay communication control method oriented to UAV cluster applications in this embodiment.

Figure 3 is a schematic diagram of the signal propagation principle of the UAV-ground control station.

Figure 4 Direction diagram of the basic vibrator.

Fig. 5 Direction diagram of monopole antenna in polar coordinates.

Fig. 6 is a schematic diagram of the principle of antenna factors affecting received signal strength.

Fig. 7 is a schematic diagram of the principle of full-time relay communication coverage of a mission point.

Fig. 8 is a schematic diagram of the principle of dynamic planning of the relay UAV in this embodiment.

FIG. 9 is a schematic flowchart of relay deployment point allocation based on the CBBA algorithm in this embodiment.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings and specific preferred embodiments, but the protection scope of the present invention is not limited thereby.

As shown in Figure 2, the steps of the air-to-ground relay communication control method for UAV cluster applications in this embodiment include:

S01. Model the UAV communication process, construct the path loss calculation model between the UAV and the ground control station, the communication model between the UAV and the ground station, and the communication model between the UAV and the relay UAV. The channel model between;

S02. Static analysis of relay demand based on the spatial distribution of mission points: obtain the spatial distribution of mission points, judge whether the current air-to-ground communication conditions need relaying according to the obtained spatial distribution and the model constructed in step S01, and if necessary, use the least relay nodes As the target, determine the number of drones that need to be relayed and the spatial location where the drones will be relayed;

S03. Relay task planning based on cluster UAV mission timing: Obtain the timing of mission points in the cluster, and construct an objective function with the goal of maximizing the relay communication benefits of UAVs to multi-task points, based on the objective function and the required The constraint conditions are constructed to obtain the dynamic relay UAV mission decision model, and the planning results of the relay mission are obtained by solving the dynamic relay UAV mission decision model.

This embodiment aims at the long-term relay communication requirements, using the air-ground communication of small and medium UAV clusters and the relay communication planning and online cooperative communication of UAV clusters in inter-machine communication, etc. Under the conditions of path effect, platform antenna gain, and communication interference, by considering the narrow-band real-time communication requirements between the UAV and the ground station, and the broadband high-speed communication requirements during the cooperative operation between the UAVs, first based on the spatial distribution of mission points with the least relay The node is the target, determine the number of relay drones that need to be relayed and the spatial location of the relay drone access, and then aim at maximizing the relay communication revenue of the drone to the multi-mission point according to the task sequence of the cluster drone Construct the objective function, and then construct the dynamic relay UAV mission decision model. By solving the dynamic relay UAV mission decision model for relay planning, the relay UAV path planning and online path adjustment for cluster tasks can be satisfied. To achieve continuous and stable transmission of command information between UAVs and the ground and inter-machine coordination information, it can not only eliminate mission point conflicts between multi-relay UAVs, but also enable the relay required for cluster tasks of air-ground communication The number of drones is minimal.

UAVs use wireless communication, and the propagation of radio waves is easily affected by various propagation media. When electromagnetic waves propagate in the air, the energy is lost, and the power is reduced accordingly. The multipath effect will introduce inter-symbol interference in the transmitted signal, which will reduce the signal-to-noise ratio of the receiver. In addition, when there is a height difference or an angle between the sending and receiving antennas, different gain effects will be produced on the receiving power of the signal. In this embodiment, the basic modeling of UAV air-to-ground communication is firstly carried out, and the above-mentioned factors are considered during the modeling, so as to construct a channel model more in line with the characteristics of the UAV.

As shown in Figure 3, the radio signal sent by the UAV first propagates in free space and reaches the low-altitude environment. Due to the influence of buildings, mountains, leaves, etc., the signal is shadowed and scattered, so that the air-to-ground communication link is carried to additional signal loss. That is, the air-to-ground propagation signal path loss is composed of two parts, the free space propagation loss and the additional loss caused by phenomena such as shadow scattering, and the additional loss is a Gaussian distribution. The average value η of the additional loss is used in modeling in this embodiment instead of a certain time A random value for the experiment. Without considering the influence of small-scale fluctuations caused by rapid changes in the propagation environment, the average path loss of the air-to-ground channel model (unit: dB) is:

L _ξ = FSPL + η _ξ (1)

Among them, FSPL represents the free-space propagation loss between the UAV and the ground control station, ξ represents the propagation group, and the additional path loss η largely depends on the signal propagation group ξ to which it belongs.

To find the spatial expectation of the path loss between the UAV and all users at an elevation angle θ, the following rule is applied:

Among them, P(ξ,θ) represents the probability of occurrence of the ξth signal propagation group when the elevation angle is θ, and L _ξ is the signal path loss value of the ξth transmission group. This embodiment follows the assumption of two propagation groups, strictly corresponding to the LoS propagation condition and the non-line-of-sight NLoS propagation condition, therefore:

NLoS links have higher path loss than LoS links due to shadowing effects and reflections of signals by obstacles. The probability of occurrence of line-of-sight communication links is related to the elevation angle and the environment. The environment can be further divided into suburbs, urban areas, and high-density urban areas, which are characterized by parameters α and β. Therefore, the line-of-sight link probability can be regarded as a continuous function of elevation angle θ and environmental parameters α, β, which can be approximated as a Sigmod function.

The probability of occurrence of a LoS link is:

The expected path loss Λ of the UAV and the ground control station is expressed as:

Λ=P(LoS,θ)×L _LoS +(1-P(LoS,θ))×L _NLoS (5)

Among them, L _LoS and L _NLoS are the average path losses of LoS links and NLoS links, respectively.

Assuming that the position of the ground control station is (x _bs , y _bs , h _bs ), and the m-coordinate of the UAV is (x _m , y _m , h _m ), the path loss (dB) can be expressed as:

c为光波速度，η_LoS和η_NLoS分别为Los链路和NLoS链路的平均附加路径损耗。Among them, f _c is the carrier frequency of radio waves, d _mb is the distance between UAV m and the ground control station,

c is the speed of light waves, η _LoS and η _NLoS are the average additional path loss of the Los link and the NLoS link, respectively.

则：The elevation angle of UAV m to the ground control station is

but:

Among them, A=η _LoS -η _NLoS ,

That is, this embodiment constructs the calculation model of the expected path loss Λ of the UAV and the ground control station according to the above formula (7).

This embodiment further constructs the antenna gain calculation model:

无关，在垂直天线的平面内为一个圆。The radiation pattern of the antenna can reflect the radiation characteristics of the antenna. Generally, the radiation pattern of the antenna indicates the distribution of the power of the electromagnetic wave radiated by the antenna in various directions. The spatial three-dimensional direction diagram of the basic matrix and the two main surfaces (E surface and H surface) are shown in Figure 4, where (a) in Figure 4 is the three-dimensional direction diagram, (b) is the E surface direction diagram, (c) is the H plane pattern. Different from the ideal power supply antenna, the basic element has directionality; the vertical antenna directly above and directly below correspond to θ=0 and θ=π of the E-plane pattern respectively, and the radio wave radiation intensity is 0. The horizontal side direction of the antenna, corresponding to the E-plane pattern θ=π/2 and θ=3π/2, the radio wave radiation intensity is the largest; according to the H-plane pattern in Figure 4 (c), it can be seen that the radio wave radiation intensity and

Regardless, it is a circle in the plane of the vertical antenna.

In this embodiment, a monopole antenna is specifically used, and its basic pattern function is:

where F(θ) is the antenna power gain.

时的单极子天线E面方向图，该图完全对称。The monopole antenna E plane pattern obtained in the specific application example is shown in Figure 5, wherein the operating frequency is 900MHZ,

The E-plane pattern of the monopole antenna at time is completely symmetrical.

In this embodiment, two factors affecting the strength of the received signal are considered, the angle θ caused by the height difference of the transmitting and receiving antennas and the change of the attitude of the aircraft itself, which causes the forming angle of the transmitting and receiving antennas. As shown in Figure 6, since the length of the antenna is very small, θ is negligible compared with the distance of the drone, so this embodiment regards the antenna as a particle, and the angle θ is the angle between the antenna center line and the vertical line, The antenna gain formed by this angle can be obtained from the antenna pattern function or the E-plane pattern. When the transmitting and receiving antennas are not parallel and there is a certain angle η, the electromagnetic wave will be decomposed on the receiving antenna when propagating, and the power will become the original cos η. Then the total antenna gain brought by the two factors is:

Integrating the signal path loss and antenna gain, the total loss Loss (dB) of the air-to-ground channel signal is:

Loss＝Λ-log 10(F(θ) cosη) (10)

This embodiment further constructs the communication model between the UAV and the ground station:

When the flying height of the UAV is H, the height of the ground control station is set to 0, all UAVs with a horizontal distance from the ground control station have the same path loss L _th , and the path loss of UAVs with a horizontal distance less than r is less than L _th , so the UAV that meets the communication quality requirement (signal-to-noise ratio is less than SNR _th ) is within the corresponding communication radius of the ground control station. According to the above air-ground channel model and the minimum signal-to-noise ratio requirements of mission UAVs, we can get:

Wherein, A=η _LoS −η _NLOS , B=20logf _c +20log(4π/c)+η _NLOS .

According to the formula (11), the communication model between the UAV and the ground station is constructed, and according to the formula (11), the coverage radius R _bs of the ground control station to the plane with a height of H can be easily obtained.

This embodiment further constructs the channel model between the UAV and the relay UAV:

The drone-to-drone channel is mainly dominated by line-of-sight communication (LoS). Although there is multipath effect caused by ground reflection, it is negligible compared with UAV-ground channel or ground-ground channel. Considering that the UAV is flying on a plane with a similar height or the same height, the height difference is almost negligible compared with the distance, and the angle θ formed by the antenna height check is approximately 0, which has almost no effect on the received signal strength, and we get:

L _uav-uav ＝20logd _uu +20logf _c +20log(4π/c)+η _LoS (12)

Among them, d _uu is the distance between drones and drones.

In the mission of large-scale fixed-wing UAV clusters, the communication link between UAVs does not consider the occlusion caused by terrain and landform, but only considers the line-of-sight propagation. The relay drone and the task drone are on the same height plane, that is, θ=90°, and the antenna gain is 1. The communication between the drone and the drone also requires that the signal-to-noise ratio is greater than the set threshold SNR _th , The maximum communication distance R _uav between UAVs on the same plane can be obtained. The channel model between UAVs and UAVs constructed in this embodiment is:

In this embodiment, based on the model constructed above, relay requirement analysis and relay planning are further performed.

The detailed steps of the relay demand analysis based on the spatial distribution of task points in step S02 of this embodiment are as follows:

Multiple UAV clusters perform tasks successively at different mission points according to the time sequence. As shown in Figure 7, the relay UAV static coverage method is used to provide communication relay services for all mission points in the whole time period. The mission drone needs to be within the communication coverage radius of its corresponding relay drone, and the relay drone also needs to be within the communication range of the ground control station. In this embodiment, in order to realize the analysis of relay requirements, the optimization goal is to deploy the least number of relay UAVs to meet the communication requirements between all mission points and the ground control station. The decision variables are the number of relay UAVs and the corresponding relay UAVs. deployment location.

Assuming that K mission points are on a plane with a height of H, the plane coordinates are represented by x _ts =[x _ts,k ,y _ts,k ] ^T ,k=1,...,K, which are outside the communication range of the ground control station , introduce the communication between the relay UAV auxiliary mission UAV and the ground control station, initially set enough K relay UAVs, and the relay position is expressed as x _re =[x _re,m ,y _{re ,m} ] ^T ,m=1,...,K. The introduction vector u=[u _m ] _1×K represents the state of the candidate relay UAV, u _m =1 means that the mth candidate relay UAV is adopted, and u _m =0 means that the candidate relay UAV is deleted. man-machine. The adopted candidate relay UAV must be within the communication radius R _bs of the ground control station. This condition can be written as:

Among them, (x _bs , y _bs ) are the plane coordinates of the ground control station. In order to make the above formula not limit the position of the relay UAV when u _m =0, it can be further converted into:

Among them, L is a constant large enough.

It should be noted that the communication equipment of the UAV is simple and limited, and the communication radius between the aircraft is generally smaller than that of the ground control station. , the communication distance from the UAV to the ground station can be greatly increased, which is greater than the communication distance between the aircraft. Therefore, when the relay UAV is within the communication radius of the ground control station, two-way communication between the air and the ground can be realized.

The matrix α=[α _mk ] _K×K represents the association between the relay drone and the mission point, and α _mk =1 means that the m-th relay drone performs relay communication for the k-th mission point, that is, the mission point k is within the communication range of the relay drone m. This condition can be written as:

2.

In order to make the above formula valid when α _mk =0, it is further written as:

3.

Limited by the capacity of the UAV communication channel, a relay UAV can connect up to N _max task UAVs at the same time.

The goal of static coverage optimization in this embodiment is to minimize the number of relay drones, determine the location of relay drones and their association with the corresponding task area, so the following optimization problem can be formed:

Among them, C ₁ is the Boolean decision variable of this problem. C ₂ means that _the communication between the mission point and the ground control station can only be relayed by one UAV; C ₃ means the upper limit of the number of mission points relayed by the relay UAV at the same time; The adopted relay UAV is associated, C ₅ and C ₆ represent the distance requirements between the relay UAV and the associated mission point and the ground control station, L is a large enough preset constant, and can guarantee α When _mk and u _m are equal to 0, there is no constraint on the m distance of the relay drone.

The above formula (18) contains quadratic terms and binary terms, which can constitute a MINLP mixed integer nonlinear programming problem, which can be solved by using the internal point optimizer in the MOSEK optimization software package in specific application embodiments.

The detailed steps for performing relay planning based on the timing of task points in step S03 of this embodiment are as follows:

S301. Modeling of relay task timing problem

Each UAV in the cluster flies to each area to perform tasks in turn according to the assigned sub-task sequence. It is assumed that the UAV needs to return the detection information when flying in the task area, but it does not need to fly between sub-task areas. Return the detection information. When the mission drone arrives at the mission point, the mission point needs to be within the communication range. Therefore, the mission point and the ground control station require different communication time, forming their own communication time windows. It is only necessary to have a corresponding relay UAV relay communication to the mission UAV at the mission point within the time window. Based on the relay static coverage deployment point obtained in the previous step S02, with the help of the rapid mobility of the UAV, the UAV can fly to the deployment point within the time window corresponding to the static coverage deployment point as required, and a UAV can be realized Fly multiple static coverage deployment points, and relay communications to several mission points in time sequence, further reducing the number of relays required. As shown in Figure 8, there are three static deployment points, which realize the full-time relay communication coverage of eight mission points. If the relay dynamic coverage is adopted, it can be reduced to two relay UAVs. After UAV 1 arrives at deployment point 1 within the time window of static deployment point 1, it then arrives at deployment point 2 before the start of the time window of static deployment point 2.

对应需与地面控制站的通信时间窗为[O_ts,k,C_ts,k]，假定中继静态部署点n为V_n个任务点提供中继通信，对应的任务点集合记为Ta_n＝{ta_n,v|v＝1,…,V_n,1≤ta_n,v≤K}。部署点n的时间窗必须将所有对应任务点的时间窗囊括在内，其时间窗记为[O_re,n,C_re,n]，时间窗起点O_re,n为对应任务点集合Ta_n中所有时间窗起点的最早时间，时间窗终点C_re,n为Ta_n中所有任务时间窗终点的最晚时间，记为：According to the deployment method of the static relay, UAV relays are arranged for all mission points at the same time, the static deployment points of the UAV relays are obtained, and the communication time window of the static overlay deployment points of the relays is calculated. task point

The time window corresponding to the communication with the ground control station is [O _{ts, k} , C _{ts, k} ], assuming that the static deployment point n of the relay provides relay communication for V _n task points, and the corresponding set of task points is denoted as Ta _n ={ta _{n, v} |v=1, . . . , V _n , 1≤ta _{n, v} ≤K}. The time window of the deployment point n must include the time windows of all corresponding task points, and its time window is recorded as [O _re,n ,C _re,n ], and the starting point of the time window O _re,n is the set of corresponding task points Ta _n The earliest time of the starting point of all time windows in Ta n, the end point of time window C _re,n is the latest time of the end point of all task time windows in Ta _n , recorded as:

After obtaining the location and time window of each relay static coverage deployment point, the distribution relay drone traverses these relay deployment points sequentially according to the time window requirements, and the problem is then transformed into a relay deployment point allocation problem with a time window.

为中继无人机m和静态覆盖部署点的对应关系，β_mn＝1表示第n个静态覆盖部署点分配给了第m架中继无人机。L_m为分配给中继无人机m的静态覆盖部署点总数，

为第m架中继无人机的飞行路径，p_mk为中继无人机所属的静态覆盖部署点，p_mk∈Re。

为对应p_m中每个静态覆盖部署点到达时间，τ_mk表示中继无人机m到达任务点p_mk的时间，目标函数具体定义如下：When constructing the objective function, maximize the relay communication income of the UAV to the multi-mission point, comprehensively consider the income of the communication of the mission point itself, the path of the relay aircraft, the time for the relay to reach the corresponding mission point, and the energy consumption of the relay aircraft And other factors. The global objective function is assumed to be the sum of the local reward values of all relay UAVs. Assuming that at least M relay UAVs need to be dispatched to N static coverage deployment points, the set of relay UAVs is recorded as V={1 ,2,...,M}, the set of static coverage deployment points is denoted as Re={1,2,...,N}.

is the correspondence between the relay UAV m and the static coverage deployment point, β _mn =1 means that the nth static coverage deployment point is assigned to the mth relay UAV. L _m is the total number of static coverage deployment points assigned to the relay UAV m,

is the flight path of the mth relay UAV, p _mk is the static coverage deployment point to which the relay UAV belongs, p _mk ∈ Re.

In order to correspond to the arrival time of each static coverage deployment point in p _m , τ _mk represents the time when the relay UAV m reaches the mission point p _mk , and the objective function is defined as follows:

Among them, the function c _mn (β _m ,p _m ,τ _m ) is the relay communication income obtained by the m-th relay drone arriving at the corresponding relay deployment point n according to the path p _m and mission time τ _m minus the flight path total energy expenditure.

Constraints constructed in this embodiment are specifically:

(1) Each relay mission point has and can only be assigned to one relay drone, namely:

(2) Each relay UAV is assigned at most L _max relay task points, which is:

(3) The relay UAV must leave after the end point O _re,n1 of the last static deployment point p _mk time window on the path, and arrive before the start point of the next static deployment point p _mk+1 time window. The flightable path within the time difference must be greater than the distance between the two relay deployment points:

Among them, v is the flight speed of the relay UAV.

Combining the above objective functions and constraints, the dynamic relay UAV task decision model is constructed as follows:

Among them, x _re,n1 and y _re,n1 represent the horizontal and vertical coordinates of relay static deployment point n1 respectively, x _re,n2 and y _re,n2 represent the horizontal and vertical coordinates of relay static deployment point n2 respectively, O _{re ,n1} , O _re,n2 represent the end points of the clock windows of relay static deployment points n1 and n2 respectively, and n1 and n2 correspond to p _mk and p _mk+1 .

S302. Solving the dynamic relay planning model.

In this embodiment, the Consensus-based Bundle Algorithm (CBBA) is used to solve the task decision model of the dynamic relay UAV, and the two stages of task bundle construction and conflict resolution are continuously iteratively executed during the solution process. until all assignments are completed. For the multi-task assignment problem, the CBBA algorithm reduces the complexity of multi-agent cooperative task assignment problem, not only can avoid conflicts, but also has the advantage of being fast and robust. In this embodiment, the distribution problem of relay deployment points is solved by using the CBBA algorithm, and the relay static coverage deployment points calculated by step S02 are assigned to each dynamic relay drone as the relay task points to be assigned in the CBBA algorithm. The allocation complexity can be greatly reduced, the allocation efficiency can be improved, and conflicts can be avoided at the same time.

As shown in Figure 9, the CBBA algorithm of this embodiment is continuously iterated by the two stages of task bundle construction and conflict elimination. The first stage is the construction of task bundles. In addition to creating a task bundle y, each relay UAV stores itself In addition to the assigned relay task points, the execution sequence of relay task points, and the task execution time, it also saves the total number of assigners corresponding to all tasks (that is, the successful bidder of the task) and the benefits that the task will bring (the winning bid value of the task). Five variables, the five variables are constantly updated with the algorithm iteration; in the second stage, the conflict is eliminated, and the adjacent relay UAVs communicate with each other to compare the list of successful bidders and the list of successful bid values stored in each, according to the time stamp The list is updated successively to resolve conflicts. These two processes are iterated repeatedly until all tasks are assigned.

The detailed steps of the task bundle construction phase in this embodiment include:

The relay drone continues to add relay task points to the task bundle until it reaches the limit of the maximum number of relay task points L _max of the relay drone or there is no free relay point, which cannot accommodate other tasks. Each relay UAV needs to maintain three lists: task bundle list b _m ={b _mk |b _mk ∈{1,...,N},k≤L _max }, path task point list p _m ={p _mk |p _mk ∈{1,…,N}, k≤L _max }, and the arrival time list of path task points τ _m ＝{τ _mk |k≤L _max }, b _m represents the middle The successor task points are numbered and arranged in the order of time when the relay task points are added to the task bundle. p _m stores the route points of the relay m, that is, the numbers of the relay task points, and are arranged in the order of the flight.

Where the relay task point n newly added to the task bundle b _m should be inserted into the corresponding p _m path, it is necessary to calculate the revenue increase that will be brought by the relay communication task at each insertion position, and the position with the largest revenue increase That is, the path sequence corresponding to the new relay task point n. In this embodiment, according to each list, the revenue increase that will be brought by relaying the communication task at each insertion position is calculated respectively, and the position with the largest revenue increase is used as the path sequence corresponding to the new relay task point n.

表示按照路径p_m顺序执行中继任务将会带来的总收益，中继任务点n加入到中继无人机m的任务束b_m中带来的收益增加量即可表示为：use

Indicates the total revenue that will be brought by executing the relay task according to the sequence of the path p _m . The increase in revenue brought by the relay task point n added to the task bundle b _m of the relay UAV m can be expressed as:

表示将新中继任务点n插入路径p_m上第δ个路径点前的总体收益。Among them, |p _m | is the number of task points assigned to the relay UAV,

Indicates the overall benefit of inserting the new relay task point n before the δth path point on the path p _m .

Then the total revenue of the relay UAV m performing tasks along p _m is expressed as:

中继无人机需在时间窗内到达任务点，到达时间窗时间τ_mk离时间窗起点

越近，则收益越大；第二个因素是任务点本身的中继通信价值Val_mk，它与中继通信任务量有关，设为和中继任务点b_mk覆盖的所有任务点通信时间之和

成正相关；第三个因素是一个惩罚项，无人机m从上一个中继任务点b_m(k-1)到达当前任务点b_mk所需要消耗的燃料，并与两任务点间的距离成正比。Among them, q ₀ is the coordinates of the starting point of the relay UAV, and v ₀ is the cruising speed of the relay UAV. s _mk (τ _mk , b _mk ) is the income obtained when the relay UAV m reaches the relay mission point b _mk at the time τ _mk , which is determined by three factors: the first factor is the time to reach the mission point τ _mk , the time window corresponding to relay task point b _mk is

The relay UAV needs to arrive at the mission point within the time window, and the arrival time window time τ _mk is far from the starting point of the time window

The closer it is, the greater the benefit; the second factor is the relay communication value Val _mk of the mission point itself, which is related to the amount of relay communication tasks, and it is set to be between the communication time of all mission points covered by the relay mission point b _mk and

is positively correlated; the third factor is a penalty item, the fuel consumed by UAV m from the previous relay mission point b _m(k-1) to the current mission point b _mk , and the distance between the two mission points Proportional.

和

分别为中继任务点b_mk以及b_m(k-1)的位置坐标，

为中继任务点b_mk中继服务的所有任务点通信时间之和，

为中继任务点b_mk中继通信

时间所带来的价值，即

Among them, _λ1 is the profit coefficient related to the time when the relay UAV reaches the relay mission point, which is a positive value, and γ is the energy consumed per unit distance,

and

are the position coordinates of relay task points b _mk and b _m(k-1) respectively,

The sum of the communication time of all task points serving relay task point b _mk relay,

Relay communication for relay task point b _mk

The value brought by time, that is

以及中标者出价向量

在具体应用实施例中，任务束构建的具体算法如下表所示。In the task bundle update process, in addition to storing the task bundle, path task point column and path task point time list, each UAV also includes two vectors, the relay task point winning bidder vector

and the winning bidder bid vector

In a specific application embodiment, the specific algorithm for task bundle construction is shown in the following table.

算法输入t-1次迭代向量值，输出为更新后第t次迭代更新后的向量，h_m存储着上次迭代中继无人机中标的中继任务点，当任务束的长度小于无人机中继m所能分配的最大任务点数目L_max并且在上次迭代存在中标的任务

则不断迭代算法(6-14行)。The above algorithm describes an iterative process of the mth relay UAV in the task bundle construction phase, how the relay UAV adds tasks to their respective task bundles and updates the five vectors maintained, where the initial moment, each intermediate Following the five vectors b _m (0), p _m (0), τ _m (0), z _m (0), and y _m (0) maintained by the UAV are all assigned empty sets,

The algorithm inputs the t-1 iteration vector value, and the output is the updated vector of the t-th iteration after the update. h _m stores the relay task point that the relay UAV won the bid in the last iteration. When the length of the task bundle is less than the unmanned The maximum number of task points L _max that can be allocated by machine relay m and there is a winning task in the last iteration

The algorithm is iterated continuously (lines 6-14).

The detailed steps of the conflict elimination phase in this embodiment are:

The relay UAV adopts a consistency strategy to converge the relay task winning bid list, and assigns the relay mission points to be reached for the relay UAV according to the bid winning list, so as to give the task UAVs at the corresponding task points of the relay mission points. Communication relay service. The consistency strategy does not need to limit the network to a specific structure, but can melt the contradictions of all tasks and achieve the goal of consistency.

因为通信链路为无向的，则对应

在冲突消除阶段，无人机中继间需要实时交流三个向量，两个向量在中继任务束创建阶段有使用，中标者向量z_m和中标者出价向量y_m，另外一个向量e_m∈R^M记录中继无人机间最近一次交换信息的时间戳，表示每个中继无人机从其他中继无人机处最近更新信息的时间。当中继无人机m₁和中继m₂间有直连链路时，即

时，中继m₁与中继m₂最近一次通信时间则为消息接收时间t_r。当两者之间无直连通路时，找到中继无人机m₁直连的所有其余中继无人机

再在这些中继无人机集合C中找到最近的时间戳，即：In this embodiment, a symmetric adjacency matrix G(τ) is used to represent the undirected graph of relay UAV communication. If there is a communication link between relay UAV m ₁ and relay UAV m ₂ at time t ,but

Since the communication link is undirected, the corresponding

In the stage of conflict elimination, three vectors need to be communicated in real time between UAV relays, two vectors are used in the relay task bundle creation stage, the successful bidder vector z _m and the successful bidder bid vector y _m , and the other vector e _m ∈ R ^M records the timestamp of the latest exchange of information between relay UAVs, indicating the time when each relay UAV updated information from other relay UAVs. When there is a direct link between the relay UAV m ₁ and the relay m ₂ , that is

, the latest communication time between relay m ₁ and relay m ₂ is the message receiving time t _r . When there is no direct connection between the two, find all the remaining relay drones that are directly connected to the relay drone m ₁

Then find the latest timestamp in the set C of these relay drones, namely:

和中标出价向量

融合更新。以中继任务点n为例，中继m₁可能执行三个动作：When the relay drone m ₁ updates information from the relay m ₂ , the successful bidder vector stored in the relay drone m ₁ needs to be

and the winning bid vector

Fusion update. Taking relay task point n as an example, relay m ₁ may perform three actions:

1) The winning bid vector is updated,

2) The winning bid vector is reset,

3) The winning bid vector remains unchanged,

和

赋予和中继m₂同样的值。当中继无人机m₁认为任务点n中标者是自己，中继m₂也认为中标者是中继i或者空，则不执行任何操作，中继m₁的中标向量保持不变。其余情况，两中继无人机存储的中标者产生冲突，则对中继m₁两个中标向量进行重置。If the winning bid for relay mission point n stored by relay UAV m ₂ is higher than that of relay m ₁ , or the successful bidder for relay mission point n stored by one of the two relays is m, m≠m ₁ ∩m≠m ₂ , and the time stamp of the latest message received by relay m ₂ on relay m is later than relay m ₁ , then relay drone m ₁ performs the bid vector update operation, and relay m ₁ of

and

Assign the same value as relay _m2 . When the relay UAV m ₁ thinks that the winning bidder of mission point n is itself, and the relay m ₂ also thinks that the winning bidder is relay i or empty, no operation is performed, and the winning vector of relay m ₁ remains unchanged. In other cases, if the successful bidders stored by the two relay UAVs conflict, the two successful bid vectors of the relay _m1 will be reset.

Assuming that there is a relay task point n in the task bundle of the relay UAV m, and there are other relays whose bids are higher than m, then m will give up the task. Or, in the consistency conflict resolution phase, the two relay UAVs communicate and negotiate to update or reset the task point n in the task bundle, and all tasks added after this task need to be released. Since the income gain generated by the task is related to its insertion position on the task path p _m , the tasks added to the m task bundle after this task will become invalid, and the corresponding vector elements of the successful bidder and the bid vector elements of the successful bidder are both reset. Firstly, it is necessary to find out the position δ _mn of the task point n in the task beam b _m of the relay UAV m:

Then reset the tasks after the δ _mnth item in the task bundle b _m :

Among them, _bin is the nth item of the task bundle of the relay UAV i.

Some embodiments of the present invention aim at the data transmission requirements between the UAV cluster mission point and the ground control station, by first modeling the communication environment of the UAV, and establishing a channel model suitable for the characteristics of the UAV platform; and then using the relay UAVs provide fixed-area communication coverage for all mission points, optimize the number of relay UAVs and the deployment locations of their corresponding static coverage; and then use the timing of mission points, that is, different mission points have different execution times, which are different from ground control stations. The required communication time is different, and the relay dynamic coverage method is used to successively reach multiple relay static coverage points, and the distribution of relay task points and the path of the relay UAV can be planned, which can quickly and efficiently realize the deployment of the air-to-ground relay UAV Planning can not only effectively eliminate mission point conflicts between multi-relay UAVs, but also minimize the number of relay UAVs required for air-ground communication cluster tasks, ensuring the command information and inter-machine coordination between UAVs and the ground Continuous and stable transmission of information.

The above are only preferred embodiments of the present invention, and do not limit the present invention in any form. Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments according to the technical essence of the present invention shall fall within the protection scope of the technical solution of the present invention.

Claims (10)

1. A kind of air-to-ground relay communication control method for unmanned aerial vehicle cluster application, it is characterized in that, the step comprises: S01. Model the UAV communication process, construct the path loss calculation model between the UAV and the ground control station, the communication model between the UAV and the ground station, and the communication model between the UAV and the relay UAV. The channel model between; S02. Static analysis of relay demand based on the spatial distribution of mission points: obtain the spatial distribution of mission points, judge whether the current air-to-ground communication conditions need to be relayed according to the obtained spatial distribution and the model constructed in step S01, and if necessary, use the minimum The relay node is used as the target to determine the number of relay UAVs and the spatial location of the relay UAV access; S03. Planning the relay task based on the task sequence of the cluster UAV: obtain the sequence of the task points in the cluster, and construct an objective function with the goal of maximizing the relay communication revenue of the drone to the multi-task point, based on the objective function and the obtained The required constraint conditions are constructed to obtain a dynamic relay UAV mission decision model, and the planning result of the relay mission is obtained by solving the dynamic relay UAV mission decision model. 2. The air-to-ground relay communication control method for UAV cluster applications according to claim 1, characterized in that, in the step S1, the ground control station position is (x _bs , y _bs , h _bs ), without The coordinates of man-machine m are (x _m , y _m , h _m ), and the elevation angle of UAV m to the ground control station is

Construct the expected path loss Λ calculation model of UAV and ground control station as:

Among them, A=η _LoS -η _NLoS ,

_ηLoS and _ηNLoS are the additional path losses of the line-of-sight communication link LoS and the non-line-of-sight communication link NLoS respectively, f _c is the carrier frequency of radio waves, d _mb is the distance between the UAV m and the ground control station,

c is the speed of light wave, α, β are environmental parameters; The total antenna gain calculation expression is:

Among them, κ is the total antenna gain, F(θ) is the antenna power gain, θ is the angle brought by the height difference of the transmitting and receiving antenna, and η is the angle formed by the transmitting and receiving antenna due to the change of the UAV's own attitude; According to the signal path loss and antenna gain, the calculation expression of the total loss Loss of the air-to-ground channel signal is constructed as follows: Loss=Λ-log 10(F(θ)·cosη). 3. the air-ground relay communication control method facing unmanned aerial vehicle cluster application according to claim 2, is characterized in that, in described step S01, the communication model between the unmanned aerial vehicle of construction and ground station is:

Wherein, A=η _LoS -η _NLOS , B=20log f _c +20log(4π/c)+η _NLOS , and SNR _th is the preset signal-to-noise ratio threshold; The channel model between the unmanned aerial vehicle and the relay unmanned aerial vehicle is:

4. The air-to-ground relay communication control method for unmanned aerial vehicle cluster applications according to claim 1, characterized in that, in the step S02, the optimization goal is to deploy the least relay unmanned aerial vehicles to meet the requirements of all mission points and ground Communication requirements between control stations, the decision variable is the number of relay drones and the corresponding relay deployment location, by minimizing the number of relay drones as the static coverage optimization goal, the location of the relay drone is determined And its association with the corresponding task area, the optimization problem formed is specifically:

Among them, C ₁ is the Boolean decision variable of the current problem, C ₂ indicates that the communication between the mission point and the ground control station can only be relayed by one UAV, and C ₃ indicates the number of mission points relayed by the relay UAV at the same time In the current upper bound, C ₄ restricts the mission point not to be associated with the unused relay UAV, C ₅ and C ₆ represent the distance requirements between the relay UAV and the associated mission point and the ground control station, L is When the constant is preset and can make α _mk and u _m equal to 0, there is no constraint on the m distance of the relay UAV, x _re,m , y _re,m represent the horizontal and vertical coordinates of the relay position, u=[u _m ] _1×K represents the state set of candidate relay UAVs, u _m = 1 means that the mth candidate relay UAV is adopted, u _m = 0 means delete the candidate UAV, R _bs Indicates the communication radius, α=[α _mk ] _K×K represents the correlation matrix between the relay UAV and the mission point, α _mk =1 means that the m-th relay UAV performs relay communication for the k-th mission point , that is, the mission point k is within the communication range of the relay UAV m, N _max represents the maximum number of mission UAVs connected to a relay UAV at the same time, x _ts,k ,y _ts,k represents the kth The horizontal and vertical coordinates of a task point. 5. The air-to-ground relay communication control method for unmanned aerial vehicle cluster applications according to claim 1, characterized in that, in the step S03, by deploying the method according to the static relay, all mission points are simultaneously arranged with unmanned machine relay to obtain the UAV relay static deployment point; calculate the communication time window of the relay static deployment point, assign the relay UAV to traverse the relay static deployment point in turn according to the time window requirements, and integrate the task The revenue of the communication point itself, the path of the relay drone, the time for the relay drone to reach the corresponding mission point, and the energy consumption factor of the relay drone to maximize the relay communication of the drone to the multi-mission point Yield constructs the objective function for the objective. 6. the air-to-ground relay communication control method facing unmanned aerial vehicle cluster application according to claim 5, is characterized in that, the objective function that aims at building the relay communication revenue of multi-task point with maximizing unmanned aerial vehicle is:

Among them, the function c _mn (β _m ,p _m ,τ _m ) is the relay communication income obtained by the m-th relay drone arriving at the corresponding relay deployment point n according to the path p _m and mission time τ _m minus the flight path total energy consumption,

is the corresponding relationship between the relay UAV m and the static coverage deployment point, β _mn = 1 means that the nth static coverage deployment point is assigned to the mth relay UAV, and L _m is the distribution to the relay UAV The total number of static coverage deployment points for m,

is the flight path of the mth relay UAV, p _mk is the static coverage deployment point to which the relay UAV belongs,

To correspond to the arrival time of each static coverage deployment point in p _m , τ _mk represents the time when the relay UAV m reaches the mission point p _mk ; The constraints include:

τ _mk ≤O _re,n1 ,τ _m(k+1) ≤O _re,n2 , (O _re,n2 -C _re,n1 )*v≥||(x _re,n1 -x _re,n2 ,y _re,n1 -y _re,n2 )|| ₂ , p _mk ＝n1,p _mk+1 ＝n2, Among them, the starting point of the time window O _re,n is the earliest time of the starting point of all time windows in the corresponding task point set Ta _n , and the end point of the time window C _re,n is the latest time of the ending time of all task time windows in Ta _n , indicating the relay position The horizontal and vertical coordinates of . 7. According to the air-to-ground relay communication control method for UAV cluster applications according to any one of claims 1 to 6, it is characterized in that, in the step S03, the bundle set algorithm CBBA based on consistency is used to The above dynamic relay UAV task decision-making model is solved. During the solution process, the two stages of task bundle construction and conflict resolution are continuously iteratively executed until all task assignments are completed. In the stage of task bundle construction, each relay has no The man-machine creates a task bundle y to store the relay task points assigned by itself, the execution sequence of the relay task points, and the task execution time, and saves the assigners corresponding to all tasks and the benefits that the tasks will bring, and in the iterative process Continuous update; in the stage of conflict elimination, adjacent relay UAVs communicate with each other to compare their respective stored bid winner lists and bid-win value lists, and update the bid-win value lists according to the sequence of time stamps. 8. The air-to-ground relay communication control method for unmanned aerial vehicle cluster applications according to claim 7, characterized in that, in the stage of the task bundle construction, the relay unmanned aerial vehicle constantly adds the relay task point to the task In the beam, until reaching the limit of the maximum number of relay task points L _max of the relay drone or there is no free relay point, each relay drone maintains the list of task bundles, the list of path task points, and the arrival of path task points Time list; according to each list, calculate the revenue increase that will be brought by the relay communication task at each insertion position, and use the position with the largest revenue increase as the path sequence corresponding to the new relay task point n, which is calculated according to the following formula The revenue increase brought by the addition of the relay task point n to the task bundle b _m of the relay drone m:

Among them, |p _m | is the number of task points assigned to the relay UAV,

Indicates the overall revenue of inserting the new relay task point n before the δth path point on the path p _m ,

Indicates the total revenue that will be brought by executing the relay tasks in sequence according to the path p _m ; The expression for calculating the total revenue of the relay UAV m performing tasks along p _m is:

Among them, q ₀ is the coordinates of the starting point of the relay UAV, v ₀ is the cruising speed of the relay UAV, s _mk (τ _mk ,b _mk ) is the arrival of the relay UAV m at the relay mission point at time τ _mk The income obtained when b _mk ; s _mk (τ _mk ,b _mk ) is determined by three factors, the first factor is the time to reach the mission point τ _mk , the second factor is the relay communication value Val _mk of the mission point itself , the sum of the communication time of the relay communication value Val _mk and all task points covered by the relay task point b _mk

is positively correlated, the third factor is a penalty item, so that the distance between the fuel consumed by the UAV m from the last relay mission point b _m(k-1) to the current mission point b _mk and the two mission points Proportional. 9. The air-to-ground relay communication control method for unmanned aerial vehicle cluster applications according to claim 7, wherein, in the stage of the conflict elimination, the relay unmanned aerial vehicle adopts a consistent strategy to converge the relay task winning list , according to the bid-winning list, assign the relay mission point to be reached to the relay UAV, so as to provide communication relay service to the mission UAV at the mission point corresponding to the relay mission point; among them, the relay UAV m ₁ and the middle When there is a direct link between m ₂ , that is

The latest communication time between relay m ₁ and relay m ₂ is the message receiving time t _r ; when there is no direct connection between the two, search for all other relay drones directly connected to relay drone m ₁

Find the most recent timestamp in the set of relay drones found C

When relay drone m ₁ updates information from relay m ₂ , the successful bidder vector stored in relay drone m ₁ will be

and the winning bid vector

Fusion update. 10. The air-to-ground relay communication control method for unmanned aerial vehicle cluster applications according to claim 7, wherein, in the stage of the conflict elimination, if the relay task point n stored in the relay unmanned aerial vehicle _m2 is higher than that of relay m ₁ , or one of the two relays stores the successful bidder for relay task point n as m, m≠m ₁ ∩m≠m ₂ , and the last time relay m ₂ received If the time stamp of the message about relay m is later than relay m ₁ , the relay UAV m ₁ performs the bid vector update operation, and assigns the bid vector and bidder vector of relay m ₂ to relay m The winning bid vector of ₁

and the winning bidder vector

When the relay UAV m ₁ judges that the successful bidder of mission point n is itself, and the relay m ₂ judges that the successful bidder is relay i or empty, no operation is performed, and the winning vector of relay m ₁ remains unchanged. When the successful bidder stored by the relay UAV conflicts, the two successful bid vectors of the relay _m1 are reset.

Images