CN112788699B

CN112788699B - Method and system for determining network topology of self-organizing network

Info

Publication number: CN112788699B
Application number: CN202011617131.6A
Authority: CN
Inventors: 田辉; 王雯; 崔雅娟; 范绍帅; 聂高峰
Original assignee: Beijing University of Posts and Telecommunications
Current assignee: Beijing University of Posts and Telecommunications
Priority date: 2020-12-30
Filing date: 2020-12-30
Publication date: 2022-07-12
Anticipated expiration: 2040-12-30
Also published as: CN112788699A

Abstract

The embodiment of the invention provides a method and a system for determining a network topology of a flight self-organizing network, wherein the method considers a routing mechanism, a safety constraint condition communication constraint condition and a measurement function for determining the network performance of the flight self-organizing network, and uses a particle swarm optimization algorithm to obtain an initial optimal deployment position of a relay unmanned aerial vehicle so as to complete the optimization of the current position of the relay unmanned aerial vehicle and determine the initial network topology of the flight self-organizing network. Therefore, the initial network topology of the self-organizing flying network is not constrained by the preset initial network topology of the self-organizing flying network, the obtained initial network topology of the self-organizing flying network better meets the requirement of the actual self-organizing flying network, and the method has universality and feasibility.

Description

Method and system for determining network topology of self-organizing network

Technical Field

The invention relates to the technical field of wireless communication, in particular to a method and a system for determining network topology of a flying ad hoc network.

Background

The flying self-organizing network can easily and quickly deploy the unmanned aerial vehicle to deal with the node faults, and the network coverage is expanded while the stability and the robustness of the network topology are guaranteed. This requires that the flying ad hoc network construct a good communication topology to ensure that a reliable communication link is established between each task-performing drone and the ground control station at which its respective task director is located. Because the mobility of the task unmanned aerial vehicle is highly dependent on a given task, and the movement model of the task unmanned aerial vehicle is determined by the task purpose and the task property, the dynamic network topology management of the flying self-organizing network needs to be realized by controlling the relay unmanned aerial vehicle, and the overall performance of the network is improved.

When the common task unmanned aerial vehicle does not execute the task at the beginning, the flying ad hoc network is not established, and the unmanned aerial vehicles are scattered in the air, so that the network topology of the flying ad hoc network is not formed. When the unmanned aerial vehicle needs to be tasked to execute the task, the related technology gives the initial network topology of the flying ad hoc network in advance, and the task is executed according to the given initial network topology of the flying ad hoc network in advance. However, the performance of the flying ad-hoc network is highly dependent on the initial network topology of the flying ad-hoc network, and can only be adjusted subsequently on the initial network topology of the flying ad-hoc network. However, the initial network topology of the predetermined flying ad hoc network may not meet the requirement of the actual flying ad hoc network.

Disclosure of Invention

The embodiment of the invention aims to provide a method and a system for determining a network topology of a flying ad hoc network, which are used for enabling the initial network topology of the flying ad hoc network to meet the requirement of an actual flying ad hoc network. The specific technical scheme is as follows:

in a first aspect, an embodiment of the present invention provides a method for determining a network topology of a flying ad hoc network, where the method includes:

acquiring a safety constraint condition and a communication constraint condition between end to end of each node in an unmanned aerial vehicle system; the safety constraint condition is that each node in each routing path is in an effective communication range, and the communication constraint condition is that the distance between each node is greater than or equal to the minimum safety distance;

acquiring a routing mechanism of a routing path between each unmanned aerial vehicle and a ground control station corresponding to each unmanned aerial vehicle;

determining a metric function for evaluating the network performance of the flying ad hoc network based on the longest link distance between links of all routing paths in the routing mechanism; the self-organizing network is formed for each unmanned aerial vehicle in the unmanned aerial vehicle system according to the position, the number and the task of each unmanned aerial vehicle in the unmanned aerial vehicle system and the routing mechanism;

constructing an optimization problem of the network topology of the flying ad hoc network through the safety constraint condition, the communication constraint condition and the performance measurement function;

and solving the optimization problem of the flight ad hoc network topology through a particle swarm optimization algorithm based on the current position of the relay unmanned aerial vehicle, the current position of the task unmanned aerial vehicle and the flight data of the unmanned aerial vehicle to obtain the initial optimal deployment position of the relay unmanned aerial vehicle so as to complete the optimization of the current position of the relay unmanned aerial vehicle and determine the initial network topology of the flight ad hoc network.

In a second aspect, an embodiment of the present invention provides a system for determining a network topology of a flying ad hoc network, where the system includes:

the first acquisition module is used for acquiring a safety constraint condition and a communication constraint condition between end to end of each node in the unmanned aerial vehicle system; the safety constraint condition is that each node in each routing path is in an effective communication range, and the communication constraint condition is that the distance between each node is greater than or equal to the minimum safety distance;

the second acquisition module is used for acquiring a routing mechanism of a routing path between each unmanned aerial vehicle and the ground control station corresponding to each unmanned aerial vehicle;

a first processing module, configured to determine, based on a longest link distance between links of all routing paths in the routing mechanism, a metric function for evaluating network performance of a flying ad hoc network; the self-organizing network is formed for each unmanned aerial vehicle in the unmanned aerial vehicle system according to the position, the number and the task of each unmanned aerial vehicle in the unmanned aerial vehicle system and the routing mechanism;

the construction module is used for constructing an optimization problem of the flying ad hoc network topology through the safety constraint condition, the communication constraint condition and the performance measurement function;

and the second processing module is used for solving the optimization problem of the network topology of the flight ad hoc network through a particle swarm optimization algorithm based on the current position of the relay unmanned aerial vehicle, the current position of the task unmanned aerial vehicle and the flight data of the unmanned aerial vehicle so as to obtain the initial optimal deployment position of the relay unmanned aerial vehicle, so that the current position of the relay unmanned aerial vehicle is optimized, and the initial network topology of the flight ad hoc network is determined.

The embodiment of the invention has the following beneficial effects:

according to the method and the system for determining the network topology of the flight self-organizing network, provided by the embodiment of the invention, the initial optimal deployment position of the relay unmanned aerial vehicle is obtained by considering a routing mechanism, a safety constraint condition, a communication constraint condition and a measurement function for determining the network performance of the flight self-organizing network and using a particle swarm optimization algorithm, so that the current position of the relay unmanned aerial vehicle is optimized, and the initial network topology of the flight self-organizing network is determined. Therefore, the initial network topology of the self-organizing flying network is not constrained by the preset initial network topology of the self-organizing flying network, the obtained initial network topology of the self-organizing flying network better meets the requirement of the actual self-organizing flying network, and the method has universality and feasibility.

Of course, not all of the advantages described above need to be achieved at the same time in the practice of any one product or method of the invention.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a first flowchart of a method for determining a topology of a network of a flying ad hoc network according to an embodiment of the present invention;

fig. 2 is a second flowchart of a method for determining a topology of a self-organizing network according to an embodiment of the present invention;

fig. 3 is a schematic flowchart of an implementation process of a topology construction method in the method for determining a topology of a network of a flying ad hoc network according to an embodiment of the present invention;

fig. 4 is a schematic flow chart illustrating an implementation process of a topology adjustment method in a method for determining a topology of a self-organizing network according to an embodiment of the present invention;

fig. 5 is a schematic flow chart illustrating an implementation process of a comprehensive topology management method in the method for determining a topology of a network of a self-organizing flying network according to an embodiment of the present invention;

FIG. 6(a) is a comparison graph of performance index values when the topology construction method shown in FIG. 3 of the present invention is applied to perform topology construction with the topology construction method in the prior art;

fig. 6(b) is a comparison diagram of the longest link distance in all active route paths when topology construction is performed by applying the topology construction method shown in fig. 3 according to the embodiment of the present invention and applying the topology construction method in the prior art;

fig. 6(c) is a comparison diagram of the shortest inter-unmanned aerial vehicle distance when the topology construction method provided by the embodiment of the invention shown in fig. 3 is applied to the topology construction by applying the topology construction method in the prior art;

fig. 7(a) is a ratio histogram of performance index values when topology management is performed by applying the integrated topology management method in the flying ad hoc network according to the embodiment of the present invention shown in fig. 5 and applying the integrated topology management method in the prior art;

fig. 7(b) is a histogram of a ratio of a longest link distance when topology management is performed by applying the integrated topology management method in the self-organizing network in flight according to the embodiment of the present invention shown in fig. 5 and applying the integrated topology management method in the prior art;

fig. 7(c) is a ratio histogram of the shortest distance between the drones when topology management is performed by applying the integrated topology management method in the flying ad hoc network according to the embodiment of the present invention shown in fig. 5 and applying the integrated topology management method in the prior art;

fig. 8 is a comparison graph of time and cost when topology management is performed by applying the integrated topology management method in the self-organizing network for flight according to the embodiment of the present invention shown in fig. 5 and applying the integrated topology management method in the prior art;

fig. 9 is a schematic structural diagram of a network topology determining system of a self-organizing network according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

First, for the convenience of understanding the embodiments of the present invention, the following terms used in the embodiments of the present invention will be described.

A mission drone herein is a drone that performs certain predetermined tasks such as detection and rescue, surveillance, and patrol, etc. After the task unmanned aerial vehicle executes the preset task and acquires the information, the information is reported to the ground space station, so that a task director can master the information. Specifically, the task unmanned aerial vehicle can play a role in various application scenes of pest control, typhoon early warning, atmosphere detection, pre-earthquake early warning and post-earthquake rescue, environment monitoring and management, public safety patrol and prevention and the like in agriculture, forestry, animal husbandry and fishery by loading different operation equipment.

Generally, in the process of reporting information back to the ground space station by the task unmanned aerial vehicle, the task unmanned aerial vehicle and the ground space station are not in the communication range of the other party, so that the information needs to be transmitted to the ground space station through the assistance of other unmanned aerial vehicles. And other unmanned aerial vehicles are located between task unmanned aerial vehicle and the ground space station, and are in other unmanned aerial vehicle and task unmanned aerial vehicle's communication range, and/or other unmanned aerial vehicle and ground space station's communication range, can follow task unmanned aerial vehicle like this and forward to ground space station through other unmanned aerial vehicle. Other drones herein may be referred to as relay drones.

Based on the introduction of the above terms, a method and a system for determining a network topology of a flying ad hoc network according to an embodiment of the present invention are described below.

When the common task unmanned aerial vehicle does not execute the task at the beginning, the flying ad hoc network is not established, and the unmanned aerial vehicles are scattered in the air, so that the network topology of the flying ad hoc network is not formed. When the task unmanned aerial vehicle needs to execute the task, the related technology gives the initial network topology of the flying ad-hoc network in advance, and the task is executed according to the given initial network topology of the flying ad-hoc network in advance. However, the performance of the flying ad-hoc network is highly dependent on the initial network topology of the flying ad-hoc network, and can only be adjusted subsequently on the initial network topology of the flying ad-hoc network. However, the initial network topology of the predetermined flying ad hoc network may not meet the requirement of the actual flying ad hoc network.

The embodiment of the invention provides a method and a system for determining a network topology of a flight self-organizing network, wherein an initial optimal deployment position of a relay unmanned aerial vehicle is obtained by considering a routing mechanism, a safety constraint condition, a communication constraint condition and a measurement function for determining the network performance of the flight self-organizing network through a particle swarm optimization algorithm, so that the current position of the relay unmanned aerial vehicle is optimized, and the initial network topology of the flight self-organizing network is determined. Therefore, the initial network topology of the self-organizing flying network is not constrained by the preset initial network topology of the self-organizing flying network, the obtained initial network topology of the self-organizing flying network better meets the requirement of the actual self-organizing flying network, and the method has universality and feasibility.

The following provides a detailed description of a method for determining a network topology of a flying ad hoc network according to an embodiment of the present invention.

The method for determining the network topology of the self-organizing network in flight is applied to an unmanned aerial vehicle system, wherein the unmanned aerial vehicle system comprises a relay unmanned aerial vehicle, a task unmanned aerial vehicle and a ground control station.

In order to support the cooperative coordination of the unmanned aerial vehicle cluster and guarantee the flexibility and effectiveness of the unmanned aerial vehicle system, the embodiment of the invention considers a limited and centralized control structure. Under the control structure, the ground control station can prearrange the attributes of each unmanned aerial vehicle, such as a task unmanned aerial vehicle or a relay unmanned aerial vehicle, and the attributes do not change in the task execution process. Meanwhile, the ground control station allocates the task unmanned aerial vehicle to a given initial task, and the movement model of the task unmanned aerial vehicle is controlled by the purpose and the property of the task. The movement of the relay unmanned aerial vehicle is influenced by an airborne control system of a relay unmanned aerial vehicle body, and when the position of a task unmanned aerial vehicle cluster changes, the airborne control system makes a cooperative decision based on the internal communication of the unmanned aerial vehicle cluster to control the next time step movement of the relay unmanned aerial vehicle. The ground control station monitors the movement of the unmanned aerial vehicle cluster in the whole task execution process, but does not interfere in the autonomous decision of the unmanned aerial vehicle.

As shown in fig. 1, a method for determining a network topology of a self-organizing network in flight according to an embodiment of the present invention may include the following steps:

step 110, acquiring a safety constraint condition and a communication constraint condition between end to end of each node in the unmanned aerial vehicle system; the communication constraints include: each node in each routing path is within an effective communication range, and the security constraint conditions include: the distance between the nodes is larger than or equal to the minimum safety distance.

It should be noted that, contain ground control station and unmanned aerial vehicle in the above-mentioned unmanned aerial vehicle system, wherein according to the unmanned aerial vehicle attribute, divide into task unmanned aerial vehicle and relay unmanned aerial vehicle with unmanned aerial vehicle. And the task unmanned aerial vehicle, the relay unmanned aerial vehicle and the ground control station are respectively regarded as three nodes.

In the above steps, in order to ensure that reliable end-to-end communication can be established between all the task unmanned aerial vehicles and their corresponding ground control stations, it is necessary to ensure that, in all the routing paths where links are established, the link range of the wireless link between end-to-end terminals must not be greater than the effective communication distance, so that a safety constraint condition is required, that is, each node in each routing path is within the effective communication range.

When the unmanned aerial vehicle cluster carries out cooperative execution tasks, a certain safety distance needs to be maintained between all unmanned aerial vehicles in the whole process, and the collision of the unmanned aerial vehicle cluster can be effectively avoided. Therefore, a communication constraint that the distance between nodes is greater than or equal to a minimum safe distance is required.

And step 120, acquiring a routing mechanism of a routing path between each unmanned aerial vehicle and the ground control station corresponding to each unmanned aerial vehicle.

Step 130, determining a metric function for evaluating the network performance of the flying ad hoc network based on the longest link distance between links of all routing paths in the routing mechanism; the self-organizing flying network is formed by unmanned aerial vehicles in the unmanned aerial vehicle system according to the positions, the number and the tasks of the unmanned aerial vehicles in the unmanned aerial vehicle system and the routing mechanism.

The metric function is the performance indicator function. The term "metric" is a measure of network performance. And the smaller the value of the performance index function is, the better the network performance of the flying ad hoc network is. For all the routing paths for establishing the links, the longest link distance between the links of each routing path may be defined as the performance indicator function, i.e. the metric function.

And 140, constructing an optimization problem of the flying ad hoc network topology through the safety constraint condition, the communication constraint condition and the performance measurement function.

And 150, solving the optimization problem of the network topology of the flight ad hoc network by a particle swarm optimization algorithm based on the current position of the relay unmanned aerial vehicle, the current position of the task unmanned aerial vehicle and the flight data of the unmanned aerial vehicle to obtain the initial optimal deployment position of the relay unmanned aerial vehicle so as to complete the optimization of the current position of the relay unmanned aerial vehicle and determine the initial network topology of the flight ad hoc network.

In step 150, the flight data of the drone is used to reflect the flight of the drone. It includes unmanned aerial vehicle flight mission and unmanned aerial vehicle flying speed. In the self-organizing network, the current position, the safety constraint condition and the communication constraint condition of the relay unmanned aerial vehicle and a measurement function for evaluating the network performance of the self-organizing network are considered, the initial optimal deployment position of the relay unmanned aerial vehicle is obtained by using a topology based on a particle swarm optimization algorithm, and the zero-following construction of the initial topology is completed.

In the embodiment of the invention, the determined initial network topology of the flying self-organizing network takes the current position, the safety constraint condition and the communication constraint condition of the relay unmanned aerial vehicle and the measurement function for evaluating the network performance of the flying self-organizing network into consideration, and the initial optimal deployment position of the relay unmanned aerial vehicle is obtained by using the particle swarm optimization algorithm so as to complete the optimization of the current position of the relay unmanned aerial vehicle and determine the initial network topology of the flying self-organizing network.

It should be noted that the three node sets are respectively represented as

Wherein G is a ground control station set, G_kIs the kth node in the ground control station set, k is the serial number of the nodes in the ground control station set, g is the ground control station, M is the task unmanned aerial vehicle set, M_kIs the kth node in the task unmanned aerial vehicle set, m is the task unmanned aerial vehicle, r_kFor the kth node in the relay unmanned aerial vehicle set, r is the relay unmanned aerial vehicle, and { } is the node set, where | · | is the cardinality of the set.

On the basis of the unmanned aerial vehicle and the ground control station definition, each unmanned aerial vehicle can complete the cooperative execution task under the safety constraint condition and the communication constraint condition. In the step 110, the following formula is adopted to obtain the communication constraint condition between end to end of each node in the unmanned aerial vehicle system:

wherein said max is the largest one of the elements selected for comparison, said max being the largest one of the elements selected for comparison

Is a node

And node

A geometric distance therebetween, i.e.

Wherein | · | purple sweet₂Is Euclidean distance, k is the node sequence number of the routing path between the task unmanned aerial vehicle and the ground control station

For task unmanned plane m_kAnd the q node of the routing path between the ground control station, wherein q is the node serial number, the

For task unmanned plane m_kAnd the q +1 th node of the routing path between the ground control station, wherein q +1 is the node serial number, d_cmFor the distance threshold value of effective communication between unmanned aerial vehicles, only when the distance between the nodes is less than the distance threshold value of effective communication between the unmanned aerial vehicles, good communication can be kept between the unmanned aerial vehicle end and the end, cm is the lower corner mark of the distance threshold value of effective communication between the unmanned aerial vehicles, and is used for distinguishing with other distances.

In the step 110, the following formula is adopted to obtain the safety constraint condition between end to end of each node in the unmanned aerial vehicle system:

wherein the min is the smallest one of the compared elements, the δ (u, v) represents the geometric distance between the node u and the node v, the u is the index of the start node, the v is the index of the arrival node, and the d is the index of the arrival node_sfFor the minimum safe distance that can avoid collision between unmanned aerial vehicle, and safe distance is far less than end-to-end communication distance, sf is the lower corner mark of the minimum safe distance symbol that can avoid collision between unmanned aerial vehicle.

Based on the above-mentioned security constraint condition and communication constraint condition between end to end of each node in the unmanned aerial vehicle system, it is assumed that the routing protocol used between nodes is known, that is, whatever routing protocol is adopted, it is regarded as an arbitrary protocol expression. And inputting the positions of all the nodes into the function, so that a routing path for realizing communication between each task unmanned aerial vehicle and the corresponding ground control station can be output. Therefore, the step 120 of the embodiment of the present invention may adopt various implementation manners, and in one implementation manner, the current positions of each unmanned aerial vehicle and the ground control station are obtained, and the following protocol expression is adopted to obtain the routing mechanism of the routing path between each unmanned aerial vehicle and the ground control station corresponding to each unmanned aerial vehicle; wherein, the protocol expression is as follows:

rho is a routing path between the task unmanned aerial vehicle and a ground control station corresponding to the task unmanned aerial vehicle, { } is a node set, and x is_GIs a three-dimensional position representation of the ground control station, G is a node set of the ground control station, x_MFor three-dimensional position representation of the mission drone, M is a set of nodes, x, of the mission drone_RIs a three-dimensional position representation of the relay drone, R is a set of nodes of the relay drone,

is a symbolic representation of the mapping relationship,

represents the k task unmanned plane m as a set of ordered nodes_kAnd task unmanned plane m_kCorresponding ground control station g^(m)M is a task unmanned plane, a group of ordered nodes is composed of all unmanned planes and a ground control station, and the ordered nodes are integrated

In (1), the k-th element in the routing path is represented as

And are assembled

First element of (1)

Must be m_kLast element

Must be that

The remaining elements are relay drones, denoted as

Wherein

|. | is the cardinality of the set.

After the routing mechanism, based on the longest link distance, a penalty performance metric function capable of measuring the network topology performance is constructed by considering a suitable routing protocol, constraint conditions and performance metric function in a self-organizing network scene, so that the determined initial network topology of the self-organizing network can meet the requirement of an actual self-organizing network, and the method has universality and feasibility, and therefore, the specific implementation steps of the step 130 in the embodiment of the invention are as follows:

based on the longest link distance between links of all routing paths in the routing mechanism, an expression of the following metric function is adopted:

determining a measurement function for evaluating the network performance of the flying ad hoc network;

wherein, f is a function of the metric,

for mission unmanned plane m_kAnd q +1 th node of the routing path between the ground control stations, the parameter α representing the degree to which the link distance affects the communication quality thereof.

And inputting the positions of all nodes and the routing paths obtained according to the positions of the nodes in the measurement function, and outputting specific values for measuring the network performance. The metric function is the performance indicator function. The term "metric" is a measure of network performance. Rather, the performance indicator function is relative to each r_kE.g. of R

The smaller the value, the better the network performance of the flying ad-hoc network, where r is_kIs the kth node of the set of relay drones, R is the set of relay drones,

for relaying unmanned aerial vehicle r_kThe three-dimensional position of (a).

Because the particle swarm optimization algorithm is mostly suitable for processing the unconstrained optimization problem, the problem has inequality constraints, namely security constraints and communication constraints. Therefore, the constraint is omitted by means of a penalty objective function method in the algorithm, and the constraint optimization problem is replaced by an unconstrained optimization problem. Therefore, an unconstrained optimization problem is constructed by setting penalty constraints for security constraints and communication constraints. Therefore, constraint is omitted by a method of punishing the objective function, and the constrained optimization problem is replaced by the unconstrained optimization problem. There are multiple implementation manners in step 140, and in one implementation manner, the optimization problem of the network topology of the ad hoc flight network is constructed by using the following formula according to the security constraint condition, the communication constraint condition, and the performance metric function:

wherein the minimize is a relay unmanned aerial vehicle changed byThree-dimensional position coordinate x of_RS is the three-dimensional deployment space of the unmanned aerial vehicle,

for a preset penalty performance indicator function:

wherein x is_GIs a three-dimensional position representation of the ground control station, G is a node set of the ground control station, x_MFor three-dimensional position representation of the mission drone, M is a set of nodes, x, of the mission drone_RFor three-dimensional position representation of the relay unmanned aerial vehicle, R is a node set of the intermediate unmanned aerial vehicle, rho is a routing path between each task unmanned aerial vehicle and a ground control station corresponding to each task unmanned aerial vehicle,

for task unmanned plane m_kThe number of nodes in the routing path, | is the cardinality of the node set,

for task unmanned plane m_kK is the node serial number of the routing path between the mission unmanned aerial vehicle and the ground control station, delta is the geometric distance between the nodes,

for task unmanned plane m_kAnd a q-th node of the routing path between the ground control station,

for task unmanned plane m_kAnd a q +1 th node, d, of the routing path between the ground control station_cmFor the distance threshold value of the effective communication between the unmanned aerial vehicles, good communication can be maintained between the sender and the receiver only when the distance between the nodes is smaller than the value, and cm is the distance threshold value of the effective communication between the unmanned aerial vehiclesLower corner mark [. C]⁺Max {0, · }, the physical meaning is to choose the larger one of them, μ is the penalty parameter of the security constraint, and d_sfIn order to avoid collision between drones at a minimum safe distance which is far shorter than the end-to-end communication distance, sf is a lower subscript of a minimum safe distance symbol which can avoid collision between drones, u is a node u, v is a node v, { lambda } is a symbol of a minimum safe distance which can avoid collision between drones_m}_m∈MA penalty parameter that is a communication constraint.

Therefore, through reasonable transformation, the optimization problem of the unconstrained flight ad hoc network topology is established.

In order to solve the optimization problem of the unconstrained flight ad hoc network topology, a particle swarm optimization algorithm is adopted to solve the flight ad hoc network topology construction problem. After initializing each particle velocity and position, the values of both are iterated a number of times to find the optimal solution to the problem. Therefore, the step 150 in the embodiment of the present invention further includes:

step 1, using the relay unmanned aerial vehicle as particles, wherein all the particles form a population, and the population is distributed in a D-dimensional target search space; and the D dimension is that the position and the flying speed of any particle correspond to a D dimension vector.

And 2, initializing the speed and the position of each particle for each particle, and acquiring the optimal position of each particle and the optimal position of the global particle in each iteration.

And 3, iterating all the particles in the target search space according to a speed updating formula and a position updating formula, updating the position and the speed of each particle until an iteration termination condition is met, and obtaining the optimal global position of the particles to serve as the initial optimal deployment position of the relay unmanned aerial vehicle.

The iteration termination condition is that the iteration frequency reaches the maximum iteration frequency or the algorithm converges to a preset precision, and the preset precision can be set according to the requirements of users.

In the step 1, a population with the number P of particles is formed by all the particles and is distributed in a D-dimensional target search space. Wherein the ith granuleThe position and the flying speed of any particle, which are children, correspond to a D-dimensional vector, which is respectively expressed as: z is a radical of formula_i＝(z_i1,z_i2,....z_iD)，i＝1,2,…P，v_i＝(v_i1,v_i2,....v_iD)，i＝1,2,…P。

Wherein, z is_iIs the D-dimensional position coordinate of particle i, said z_i1Is the 1 st dimensional position coordinate of particle i, said z_i2Is the 2 nd dimensional position coordinate of particle i, said z_iDIs the D-dimensional position coordinate of the particle i, i is the serial number of the particle, D is the position/flying speed dimension of the particle, P is the number of the particles in the population, v_iIs the D-dimensional flight velocity of the particle i, v_i1Is the 1 st dimension flight velocity of the particle i, v_i2Is the 2 nd dimension flight velocity of the particle i, said v_iDIs the D-th dimension flight velocity of particle i.

And (3) mapping the unconstrained optimization problem of the flying ad hoc network topology into the optimization problem of the flying ad hoc network topology in a three-dimensional real number search space by combining the basic principle of the optimization particle swarm optimization algorithm:

wherein z is a decision variable of the optimization problem.

Obtaining the optimal position p of each particle in the step 2_iAnd the optimal position g of the global particle, as two extreme values, first, the velocity update formula in the step 3 is executed as follows:

and iterating all the particles in the target search space, and updating the position and the speed of each particle.

Wherein,

is any, e isSaid

The velocity of the ith particle at the (l + 1) th iteration; the l is iteration times and represents the ith iteration of the particle swarm optimization algorithm; w is an inertia factor, the value of which is non-negative; c is said₁And c₂Is a learning factor; to meet the requirements of time overhead and network performance, u₁,u₂Is set at [0,1 ]]ⁿThe random variables are independently and uniformly distributed so as to enhance the randomness of the influence of the historical speed of the particles on the current speed; the above-mentioned

For the optimal position of the ith particle itself at the ith iteration

Is the position of the ith particle at the ith iteration, the g^lThe optimal position of the global particle at the ith iteration.

Then, the position updating formula in the step 3 is executed as follows:

iterating all particles in the target search space, and updating the position of each particle;

wherein, the

The position of the ith particle at the (l + 1) th iteration.

And performing iterative updating on all the particles in the search space according to the speed updating formula and the position updating formula until an iteration termination condition is met. When the iteration termination condition is met, the optimal global position of the particles is a final solution obtained by the particle swarm optimization algorithm aiming at the topology construction problem, namely the optimal deployment position of the relay unmanned aerial vehicle.

In the embodiment of the invention, a particle swarm algorithm which is simple and easy to realize, less in parameter adjustment and high in convergence speed is adopted as an algorithm basis, an optimal solution meeting a topology construction problem function is sought by carrying out a series of speed updating and position updating on particles, an optimal deployment position of the relay unmanned aerial vehicle can be found on the basis of meeting the requirements of time overhead and network performance at the same time, and the construction from zero of an initial topology is completed.

In a practical scenario, the task drone needs to move continuously according to its given task, and thus the optimality of the relay drone location is also lost over time. Therefore, based on the network topology change caused by the random movement of the task unmanned aerial vehicle, the network topology can be incrementally adjusted based on the quasi-newton method, as shown in fig. 2, the specific implementation manner is as follows:

step 210, obtaining a safety constraint condition and a communication constraint condition between end to end of each node in the unmanned aerial vehicle system; the safety constraint condition is that each node in each routing path is in an effective communication range, and the communication constraint condition is that the distance between each node is greater than or equal to the minimum safety distance;

step 220, acquiring a routing mechanism of a routing path between each unmanned aerial vehicle and a ground control station corresponding to each unmanned aerial vehicle;

step 230, determining a metric function for evaluating the network performance of the flying ad hoc network based on the longest link distance between links of all routing paths in the routing mechanism; the self-organizing network is formed for each unmanned aerial vehicle in the unmanned aerial vehicle system according to the position, the number and the task of each unmanned aerial vehicle in the unmanned aerial vehicle system and the routing mechanism;

step 240, constructing an optimization problem of the flying ad hoc network topology through the safety constraint condition, the communication constraint condition and the performance measurement function;

and 250, solving the optimization problem of the network topology of the flight ad hoc network by a particle swarm optimization algorithm based on the current position of the relay unmanned aerial vehicle, the current position of the task unmanned aerial vehicle and the flight data of the unmanned aerial vehicle to obtain the initial optimal deployment position of the relay unmanned aerial vehicle so as to complete the optimization of the current position of the relay unmanned aerial vehicle and determine the initial network topology of the flight ad hoc network.

Step 260, acquiring the position of each unmanned aerial vehicle, the position of a ground control station, the movement increment of the unmanned aerial vehicle and a routing mechanism in the initial network topology of the self-organizing flying network at the current time step;

and 270, calculating the gradient of the position at each current time step by adopting a quasi-Newton method based on the positions of all the unmanned aerial vehicles, the positions of the ground control stations, the movement increment of the unmanned aerial vehicles and a routing mechanism in the initial network topology of the self-organizing flying network at the current time step, and iterating to obtain the optimal adjustment position of each relay unmanned aerial vehicle at the next time step.

Steps S210 to S250 are the same as steps S110 to S150, respectively, and the same effect can be achieved. In addition, the execution sequence of S210 to S270 is not necessarily limited to the sequence shown in fig. 2 in the embodiment of the present invention, and may be adjusted from front to back.

Therefore, due to network topology change caused by random movement of the task unmanned aerial vehicle, a quasi-Newton method with low time overhead is designed, and the network topology is subjected to incremental adjustment;

based on the establishment of the network topology and the adjustment using the quasi-newton method, in the

above steps

260 and 270, the cumulative changes of the flying ad hoc network caused by the random movement of the drone are dealt with. Therefore, the network topology is adjusted in increment based on the quasi-Newton method so as to adapt to the random movement of the mission unmanned aerial vehicle. Therefore, due to network topology change caused by random movement of the task unmanned aerial vehicle, a quasi-Newton method with low time overhead is designed, and the network topology is subjected to incremental adjustment; in a real scenario the task drone needs to move continuously according to its given task, and thus the optimality of relay drone position is lost over time. The requirements of the flight self-organizing network on effectiveness and timeliness can be combined with the requirements of an actual scene, the network topology is adjusted with low complexity, and good performance of the network is maintained.

In a specific embodiment of the invention, the performance metric function is expressed at a time step t, relative to each relay nobodyMachine r_kThe first derivative of the position of (c):

wherein,

r for each relay drone at time step t for a performance metric function_kIs a function of the first derivative of the position of (a),

for relaying unmanned aerial vehicle r_kThree-dimensional position at the time of the current time step t, r_kIs the kth relay unmanned plane, r is a relay unmanned plane node, k is the serial number of the relay unmanned plane node,

at time step t for a performance metric function, r with respect to each relay drone_kFirst derivative of the position of (1), x_GIs the three-dimensional position of the ground control station, G is the set of ground control stations, x_M(t) is the position of the task unmanned aerial vehicle set at time step t, M is the task unmanned aerial vehicle set, x_R(t) is the position of the relay unmanned aerial vehicle set at the time step t, R is the relay unmanned aerial vehicle set,

for mission unmanned plane m_kRouting path, m_kIs the k task unmanned plane.

Iteration is carried out based on a quasi-Newton method to obtain the next time step t +1 position, and each relay unmanned aerial vehicle r_kThe position of (c):

(1) initialization

D_jSetting an iteration threshold a given the parameters I, j, 0

σ∈(0,0.5)。

Wherein x is_jThe position of the relay unmanned aerial vehicle obtained in the jth iteration, j is the iteration number, g_jFor the j-th iteration the performance metric function is relative to each relay drone r at time step t_kFirst derivative of the position of, D_jThe matrix is an inverse matrix of an approximate sea plug matrix in the jth iteration, I is an identity matrix, j is the iteration times, and sigma is an auxiliary parameter.

(2) And calculating a search direction: d_j＝-D_jg_j；

Wherein d is_jIs the search direction at the jth iteration.

(3) Computing satisfaction inequality

Smallest non-negative integer b_j. Order to

Considering the speed limit of the unmanned aerial vehicle, the speed threshold value of the relay unmanned aerial vehicle is considered when determining the position movement of the relay unmanned aerial vehicle, and the following results are obtained:

wherein x is_jThe position of the relay unmanned aerial vehicle obtained in the jth iteration is obtained,

for the search displacement at the jth iteration,

for a given auxiliary parameter, b_jIs the auxiliary parameter at the j-th iteration,

for the j-th iteration the performance metric function is unmanned with respect to each relay at time step tMachine r_kIs the transposition of the first derivative of the position of (1), T is the transposition operation, x_j+1Is the position of the relay unmanned aerial vehicle delta obtained in the (j + 1) th iteration_jFor the search displacement at iteration j, if is |)₂In the form of the euclidean distance,

for relaying unmanned aerial vehicle r_kThe maximum distance moved at each time step.

(4) Calculate g_j+1＝g(x_j+1) If g | | |_j+1||₂<a, stopping the iteration,

otherwise, calculate h_j＝g_j+1-g_j，s_j＝x_j+1-x_jUpdate

Wherein, g_j+1For the (j + 1) th iteration, the performance metric function is relative to each relay drone r at a time step t_kA is an iteration threshold, h_jFor the (j + 1) th iteration, the performance metric function is relative to each relay drone r at a time step t_kWith respect to each relay drone r at time step t, the first derivative of the position of (d) and the performance metric function at the jth iteration_kOf the first derivative of the position of (a), s_jIs the difference value, x, between the position of the relay unmanned aerial vehicle obtained in the (j + 1) th iteration and the position of the relay unmanned aerial vehicle obtained in the (j) th iteration_j+1The position x of the relay unmanned aerial vehicle obtained in the j +1 th iteration_jThe position of the relay unmanned aerial vehicle obtained in the jth iteration, j is the current iteration number, and D_j+1Is the inverse of the approximate sea plug matrix at the j +1 th iteration, D_jIs the inverse of the approximate sea plug matrix at the jth iteration,

function of performance measure for j +1 th iterationNumber r at time step t with respect to each relay drone r_kWith respect to each relay drone r at time step t, the first derivative of the position of (d) and the performance metric function at the jth iteration_kIs transposed to the difference of the first derivative of the position of (a),

is the transpose of the inverse of the approximate sea plug matrix at the jth iteration.

(5) And (5) enabling j to be j +1, and turning to the step (2).

Calculating the topology edit distance of the initial network topology of the flying self-organizing network between the current time step t and the next time step tau so as to measure the change degree of the initial network topology of the flying self-organizing network from the current time step t to the next time step tau;

wherein the initial network topology of the flying ad hoc network is represented as a graph at the current time step; wherein the graph has a set of nodes, a set of edges, and a set of node locations; the node set is a set of all nodes in the flying ad hoc network, the edge set is a set of all unordered node pairs with lengths not greater than a preset length, the node position set is positions of all nodes in the node set, and the nodes include: one or more of each unmanned aerial vehicle and a ground control station;

calculating the topology edit distance between the current time step and tau of the initial network topology of the flying self-organizing network by adopting the following formula based on the graph:

wherein, delta_tedFor the topology edit distance, ted is the subscript of the topology edit distance, T (T) is the flying ad hoc network topology at the current time step T, T (τ) is the flying ad hoc network topology at the next time step τ, { ρ_mIs the set of routing paths, w_iWeight parameter for topology editing operation, e_iFor topology editing operations, t is the current timeAnd (4) an inter step length, wherein tau is the next time step length.

If the topology edit distance is smaller than a first threshold, executing the positions of all unmanned aerial vehicles, the positions of ground control stations, the movement increment of the unmanned aerial vehicles and a routing mechanism in the initial network topology of the self-organizing flying network based on the current time step, and iterating by adopting a quasi-Newton method to obtain the optimal adjustment position of each relay unmanned aerial vehicle at the next time step;

and if the topology edit distance is larger than a first threshold, executing the updating of the routing path or the reconstruction of the topology so as to reduce the cumulative change of the flying self-organizing network caused by the random movement of each unmanned aerial vehicle.

In a specific embodiment of the present invention, the topology edit distance is a weighted sum of topology edit operations. The concrete expression is as follows: modeling a flying ad-hoc network topology at time t as a graph T (t) having a set of nodes N, a set of edges ε (t), and a set of node locations X_N(t) of (d). The node set N is a set of all wireless nodes in the self-organizing network in flight, and the edge set epsilon (t) is a set of all wireless nodes with the length not greater than d_cmThe set of unordered node pairs, the set of node locations X_NAnd (t) is the position of all nodes in the set N.

Ignoring node insert and delete operations, only the minimum number of edge insert and delete operations required to convert from T (T) to T (τ) is considered. The two are taken as the first two topology edit distances, namely:

e₁(t,τ)＝|ε(τ)\ε(t)|

e₂(t,τ)＝|ε(t)\ε(τ)|

taking the sum of the side length changes as a third topology editing operation:

where δ (u, v, t) is a distance between nodes u and v at time step t, that is, δ (u, v, t) | | x_u(t)-x_v(t)||₂。

To avoid violating end-to-end communication constraints and security constraints, the algorithm should perform the operation of routing path update or topology reconfiguration when the constraints become strict. The extent to which the network topology T (τ) at time τ violates these two constraints needs to be considered. Two additional topology editing operations are defined whose values increase as the communication and security constraints become more stringent, respectively, namely:

wherein psi₁And psi₂Is a sensitivity parameter, | is the cardinality of the set of nodes,

task unmanned aerial vehicle m_kThe number of nodes in the routing path, q is the node serial number, delta is the node distance, and d_sfTo avoid collisions between drones at a minimum safe distance that is much smaller than the end-to-end communication distance, exp (-) is indexed to (-) for example.

Finally, defining the topology edit distance of the self-organizing network topology in flight between the current time step t and the next time step tau:

wherein w_iIs a topology editing operation e_iThe weight parameter of (a) is determined,

is a collection of routing paths among the currently active routing paths. The current active routing path is the routing path for which a connection has been established.

A larger topology edit distance value means that the flying ad hoc network topology has changed greatly or that the constraints have become more stringent. And calculating the topology edit distance of the self-organizing-in-flight network topology between the current time step t and the next time step tau, and determining whether to update the routing path or reconstruct the self-organizing-in-flight network topology based on the value.

Further, two thresholds e are set₁And e₂Wherein e is₁<∈₂. And adjusting the network topology by using a quasi-Newton method, comparing the updated network topology with a reference topology, and monitoring the change degree of the flying self-organizing network topology and the degree of the satisfied constraint condition. If the topology edit distance between the two does not exceed the threshold value epsilon₁If the estimated value is less than the preset threshold value, the flight self-organizing network topology is changed slightly, and the quasi-Newton method can be continuously used to adapt to the movement of the mission unmanned aerial vehicle. The reference topology is a topology used as a reference topology and used for comparison. The reference topology is also always changed, and for this time step, the reference topology is the topology of the last time step. Initially, a reference topology of the initial topology may be set to the initial topology itself.

Further, if the topology edit distance is greater than the threshold e₁If the change of the self-organizing network topology is large, the operation of updating the routing path or reconstructing the topology is needed to be executed. Since the topology reconstruction overhead is large, it is first considered to solve this situation by routing path updates. If the topology edit distance after the route path is updated is still larger than the threshold value epsilon₂Then the operation of reconstructing the flying ad hoc network topology is performed.

Referring to fig. 3, fig. 3 is a schematic diagram of a topology construction implementation flow in the method for determining a topology of a flying ad hoc network according to the embodiment of the present invention. In step 150 of the embodiment of the present invention, the optimization problem of the network topology of the ad hoc network is solved through a particle swarm optimization algorithm to obtain the initial optimal deployment position of the relay unmanned aerial vehicle, which may include the following steps:

and 301, acquiring various parameters required by the particle swarm optimization algorithm.

In a specific embodiment of the invention, the application of a flying ad-hoc network is combinedWith the scene, various parameters required by the particle swarm optimization algorithm comprise: safety distance threshold d_sfCommunication distance threshold d_cmInitial position x of task unmanned aerial vehicle_MPosition x of ground control station_GRouting mechanism rho used by system model, penalty coefficients lambda and mu, particle number N, inertia weight w and learning factor c₁And c₂And the like.

Step 302, initializing the positions and velocities of all particles in the particle swarm.

In a specific embodiment of the invention, the positions and velocities of all the particles in the population P are initialized according to the principles of the basic particle swarm optimization algorithm. Initialization position x_R,iFor searching for random values in space S, velocity v is initialized_R,iIs 0.

Step 303, obtaining the current optimal position of each particle and the routing path corresponding to the current optimal position of each particle, finding the particle providing the minimum penalty performance index value in the particle swarm, and determining the current global optimal position and the optimal routing path where the current global optimal position is located.

In an embodiment of the present invention, since each particle i is initialized for the first time, its current position is the best position of its current position

Obtaining the route path corresponding to the position according to the given route function

After the initialization of all the particles in P is completed, finding out the particle j which can provide the minimum penalty performance index value in the particle swarm, and determining the optimal position of the particle j

And its corresponding routing path

Is the current global optimum position

And its corresponding optimal routing path

And step 304, iteratively updating the speed and the position of each particle and the routing path of each particle to obtain the updated optimal position of a single particle and the updated optimal routing path of the single particle, and searching and obtaining the optimal positions of all the particles and the routing paths of all the particles in the optimal positions of all the single particles and the optimal routing paths of the single particles.

In a particular embodiment of the invention, the velocity v of each particle i is updated iteratively_R,iPosition x_R,iAnd corresponding routing path

Then updating its best position

And updating the corresponding optimal routing path

After all particles are updated, the searched best location so far and its corresponding routing path are updated to the global best location

And its corresponding routing path

The velocity v of each particle in the population P during the iterative update_R,iAnd randomly updating according to the respective optimal position and the global optimal position, wherein an updating formula in the algorithm is as follows:

wherein,

is the speed of the particle i in the (l + 1) th iteration, i is the particle serial number, R is the relay unmanned aerial vehicle set, l +1 is the (l + 1) th iteration, w is the inertia weight,

is the velocity of particle i at the first iteration, c₁And c₂Is a learning factor, u₁,u₂Is at [0,1 ]]^|3*R|Wherein the random vectors are independently and uniformly distributed, and the degree is Hadamard product,

the optimal position of the particle i at the first iteration is the optimal representation,

is the position of the particle i in the first iteration, is the first iteration of the flying self-organizing network topology building algorithm,

is the global optimal position at the ith iteration,

p is a particle swarm set, l is the l iteration of the topology construction algorithm of the flying self-organizing network, and u₁,u₂Is at [0,1 ]]^|3*R|Independent uniformly distributed random vectors. Since the velocity update of a particle is performed using the above equation, its velocity value may increase to a value beyond its search space range, especially for particles whose initial position is far from its own optimal position or global optimal position. Thus, the particle swarm optimization algorithm is prevented from diverging using the method of setting the threshold, the velocity of each particle i

Updating formulasComprises the following steps:

wherein,

for the threshold of the speed limit, V is the speed, max is the corner mark of the maximum value, and j is the element number.

Is composed of

The j element of (2), the

Is the threshold for the speed limit. After the update of the velocity is finished, the update of the particle position is performed according to the following formula:

obtaining a routing path corresponding to the optimal position of the particle i according to a routing mechanism

And 305, determining a current minimum penalty performance index value by using the current optimal positions of all the particles and the routing paths of all the particles, comparing the current minimum penalty performance index value with a penalty performance index value obtained by the optimal routing path of the current global optimal position, and updating the global optimal value.

In the embodiment of the present invention, if the updated optimal positions and the penalty performance index values obtained by the routing paths where all the particles are located are smaller, the current optimal positions of all the particles are updated to the initial optimal deployment positions of the current relay unmanned aerial vehicle. After obtaining the optimal positions of all the particles and the routing paths of all the particles, determining the optimal positions of the particles j and the routing paths of the particles j, and obtaining the current minimum penalty performance index value. And if the current minimum penalty performance index value is smaller than the penalty performance index values obtained by the current global optimum position and the optimum routing path where the current global optimum position is located, updating the current global optimum position and the routing path where the current global optimum position is located, and taking the current global optimum position and the routing path where the current global optimum position is located as the corresponding values of the particle j. If the current minimum penalty performance index value is not less than the penalty performance index value obtained by the current global optimum position and the optimum route path where the current global optimum position is located, the step 306 is executed.

And step 306, meeting a termination condition to obtain an initial optimal deployment position of the relay unmanned aerial vehicle so as to complete optimization of the current position of the relay unmanned aerial vehicle and determine the initial network topology of the flying self-organizing network. Thus, through the steps 301 to 305, the updating process is repeated until the termination condition is met, and a final solution of the topology construction problem in the self-organizing network is obtained.

In a specific embodiment of the present invention, this series of updating processes is repeated until a termination condition is met. When the end condition is met, the particle swarm optimization algorithm outputs the global optimal position

The optimal initial deployment position of the relay unmanned aerial vehicle can be obtained according to the final problem solution obtained by the method for constructing the topology of the self-organizing network.

By combining the construction of the initial network topology of the flying ad hoc network of the above embodiment, the embodiment of the present invention can also adjust the initial optimal deployment position of the relay unmanned aerial vehicle. Referring to fig. 4, fig. 4 is a schematic diagram illustrating a topology adjustment implementation flow in the method for determining a topology of a flying ad hoc network according to the embodiment of the present invention. The method in the embodiment of the invention further comprises the following steps: the method for obtaining the optimal adjustment position of each relay unmanned aerial vehicle specifically comprises the following steps:

step 401, obtaining the position of the ground control station at the last time step, the position of the task unmanned aerial vehicle at the last time step, the position of the relay unmanned aerial vehicle at the last time step, and flight data of the unmanned aerial vehicle.

Wherein, unmanned aerial vehicle flight data includes: routing mechanisms used by the system model.

And 402, calculating the gradient of the position at each current time step by adopting the positions of all unmanned aerial vehicles, the positions of the ground control stations, the movement increment of the unmanned aerial vehicles and a routing mechanism in the initial network topology of the self-organizing flying network based on the current time step, and iterating to obtain the optimal adjustment position of each relay unmanned aerial vehicle at the next time step. Therefore, the optimal adjusting position of each relay unmanned aerial vehicle at the next time step is obtained through iteration so as to adjust the topology of the flying self-organizing network.

In a particular embodiment of the invention, the performance metric function is relative to each relay drone r at a time step t_kThe first derivative of the position of (a) is:

wherein, X is_M(t) and X_RAnd (t) respectively integrating the positions of the task unmanned aerial vehicle and the relay unmanned aerial vehicle at the time step t.

Iteration is carried out based on a quasi-Newton method to obtain the next time step t +1 position, and each relay unmanned aerial vehicle r_kThe position of (2):

(1) and initialization of the system

D_jSetting an iteration threshold a given the parameters I, j, 0

σ∈(0,0.5)。

Wherein x is_jFor the position of the relay drone, g, obtained during the jth iteration_jFor the j-th iteration the performance metric function is relative to each relay drone r at time step t_kPosition ofFirst derivative of position, D_jIs the inverse matrix of the approximate sea plug matrix in the jth iteration, I is the unit matrix, and j is the iteration number.

(2) And calculating the search direction: d_j＝-D_jg_j。

(3) And the calculation satisfies the inequality

Smallest non-negative integer b_j. Order to

wherein, the

(4) Calculating g_j+1＝g(x_j+1) If g | | |_j+1||₂<a, stopping the iteration,

otherwise, calculate h_j＝g_j+1-g_j，s_k＝x_k+1-x_kUpdate, update

(5) And j is made to be j +1, and the step (2) is carried out.

In combination with the construction and/or adjustment of the initial network topology of the flying ad hoc network of the above embodiments, the embodiments of the present invention may also manage the initial network topology of the flying ad hoc network. Referring to fig. 5, fig. 5 is a schematic flow chart illustrating an implementation process of a comprehensive topology management method in the method for determining a topology of a network of a flying ad hoc network according to the embodiment of the present invention. In conjunction with the construction and adjustment of the initial network topology of the flying ad hoc network, the method in an embodiment of the present invention further includes: the method for managing the initial network topology of the flying ad hoc network specifically comprises the following steps:

step 501, acquiring the network topology of the current flying self-organizing network and the network topology of the flying self-organizing network after the last adjustment, and taking the network topology of the flying self-organizing network after the last adjustment as the reference topology of the network topology of the current flying self-organizing network. And at the beginning, the reference topology of the flying self-organizing network is the initial network topology.

In a specific embodiment of the present invention, the initial network topology of the self-organizing network in flight is obtained through steps 110 to 150, and an initial routing path in the initial network topology of the self-organizing network in flight is determined. And setting the network topology of the acquired flying ad hoc network as a reference topology.

Step 502, obtaining the optimal adjustment position of each relay unmanned aerial vehicle, monitoring the change degree of the network topology of the flying ad hoc network compared with the reference topology, and updating the routing path or reconstructing the topology according to the topology editing distance. Wherein updating the routing path or reconstructing the topology further comprises: and if the topology edit distance is smaller than a first threshold, executing the steps of iteratively obtaining the optimal adjustment position of each relay unmanned aerial vehicle at the next time step by adopting a quasi-Newton method based on the positions of all unmanned aerial vehicles, the positions of the ground control stations, the movement increment of the unmanned aerial vehicles and a routing mechanism in the initial network topology of the self-organizing flying network at the current time step, or if the topology edit distance is larger than the first threshold, executing the step of updating a routing path or reconstructing the topology so as to reduce the cumulative change of the self-organizing flying network caused by the random movement of all unmanned aerial vehicles.

In a specific embodiment of the invention, a quasi-newton based method is used to adjust the flying self-organizing network topology to adapt to the random motion of the mission unmanned aerial vehicle, and the updated network topology is compared with a reference topology to monitor the degree of change of the flying self-organizing network topology and the degree to which the constraint conditions are satisfied. If both are presentDoes not exceed a threshold e₁If the network topology is not changed, the network topology is adjusted by using the topology adjustment algorithm; if the topology edit distance is greater than the threshold e₁It indicates that the network topology has changed to a point where routing path updates or topology reconfiguration are required. Since the topology reconstruction overhead is large, it is considered first to solve this situation by routing path update. And if the topology edit distance after the routing path is updated is still larger than the threshold value epsilon₂Then the network topology needs to be reconstructed. And meanwhile, updating the reference topology into the current flying self-organizing network topology.

Based on the above, the following examples of the embodiments of the present invention are:

assuming 100 time steps, the position of the mission drone changes in each time step because the mission drone moves randomly. Aiming at the movement of the task unmanned aerial vehicle, the topology management process of the embodiment of the invention is as follows:

(1) according to the position of the task unmanned aerial vehicle in the first time step, the optimal deployment position of the relay unmanned aerial vehicle in the initial period is solved by using a particle swarm optimization algorithm, and an initial topology is constructed;

(2) setting the initial topology as a reference topology;

(3) and in the next time step, the position of the task unmanned aerial vehicle is changed. Adjusting the topology by using a topology adjustment algorithm to adapt to the change of the task unmanned aerial vehicle;

(4) calculating the edit distance between the time step and the topology in the last time step, and if the edit distance is more than the threshold value belonging to the same group as the current topology₁If the estimated variation is less than the preset threshold, the accumulated variation of the topology of the flying self-organizing network is indicated, and the quasi-Newton method can be continuously used to adapt to the movement of the mission unmanned aerial vehicle; and if the topology edit distance is greater than the threshold e₁If the change of the self-organizing network topology is large, the operation of updating the routing path or reconstructing the topology is needed to be executed. Since the topology reconstruction overhead is large, it is first considered to solve this situation by routing path updates. If the topology edit distance after the route path is updated is still larger than the threshold value epsilon₂Then executeReconstructing operations of a flying ad hoc network topology;

(5) updating the reference topology to be the current topology;

repeating (3), (4) and (5) until the end of 100 time steps.

Based on the discussion of the above specific embodiment, in order to further explain the beneficial effects of the optimization algorithm based on the particle swarm in the self-organizing network for flight provided by the embodiment of the present invention, the embodiment of the present invention compares the technical effects obtained by applying the topology construction method in the self-organizing network for flight provided by the embodiment of the present invention with the technical effects obtained by applying the topology construction method in the self-organizing network in the prior art through computer simulation. The method comprises the following specific steps:

the prior art participating in the comparison is: a topology construction method for deploying relay unmanned aerial vehicles based on optimization random in a flight self-organizing network.

For convenience of description, in a simulation experiment, a PSO is used to represent a topology construction method based on a particle swarm optimization algorithm in a self-organizing network in flight provided by the embodiment shown in fig. 2 of the present invention, and an RND is used to represent a topology construction method based on an optimized random deployment relay unmanned aerial vehicle in a self-organizing network in flight. In the figure, the RU is a relay drone.

Specifically, referring to fig. 6(a), fig. 6(b), and fig. 6(c), performance index values, longest link distances among all active routing paths, and shortest inter-drone distances when the topology construction method provided by the embodiment of the present invention shown in fig. 2 and the topology construction method in the prior art are applied to perform topology construction are respectively shown in comparison diagrams.

As can be seen from fig. 6(a), the network performance metric values of the PSO scheme are all lower than those of the RND scheme when the number of relay drones is 1-10. In other words, the network topology performance realized by the topology construction method based on the particle swarm optimization algorithm in the flying self-organizing network provided by the embodiment of the invention is better.

As can be seen from fig. 6(b), when the number of relay drones is 1 to 10, the longest link distance of the PSO scheme is lower than that of the RND scheme. In other words, the topology construction method based on the particle swarm optimization algorithm in the self-organizing network provides more reliable end-to-end communication.

As can be seen from fig. 6(c), when the number of relay drones is 1 to 10, the shortest inter-drone distances of the PSO scheme are all higher than those of the RND scheme. In other words, the topology construction method based on the particle swarm optimization algorithm in the self-organizing network can effectively prevent collision among unmanned aerial vehicles.

Based on the above discussion of the specific embodiment, to further explain the beneficial effects of the topology management method based on the particle swarm optimization algorithm in the self-organizing flying network provided by the embodiment of the present invention, the embodiment of the present invention compares the technical effect obtained by applying the comprehensive topology management method in the self-organizing flying network provided by the embodiment of the present invention with the technical effect obtained by applying the comprehensive topology management method in the self-organizing flying network in the prior art through computer simulation. The method comprises the following specific steps:

the prior art participating in the comparison is: and only executing the comprehensive topology management method based on the particle swarm optimization algorithm in the flying self-organizing network. In the method, a topology construction method based on a particle swarm optimization algorithm is executed in each time interval according to the new position of the task unmanned aerial vehicle. According to the method flow, the method is an optimal topology management scheme which can be realized by using the proposed topology construction method under the condition of not considering time overhead.

For convenience of explanation, in a simulation experiment, the integrated topology management method in the self-organizing network in flight provided by the embodiment shown in fig. 2 of the present invention is represented by IFTM, and the integrated topology management method in the prior art that Only executes a particle swarm optimization algorithm is represented by PSO-Only.

Specifically, referring to fig. 7(a), fig. 7(b), and fig. 7(c), ratio histograms of performance index values, longest link distances, and shortest inter-drone distances when topology management is performed by applying the integrated topology management method in the prior art and the integrated topology management method in the self-organizing network according to the embodiment of the present invention shown in fig. 4 are respectively shown.

A ratio of 1 in the ratio histogram indicates that both methods provide similar performance under this index. The ratio values for FIG. 7(a) have a mean value of 1.01575 and a standard deviation of 0.188434; for FIG. 7(b) there is an average value of 1.01426 with a standard deviation of 0.139331; for FIG. 7(c) there is an average value of 1.00816 and a standard deviation of 0.12636. The three subgraphs show that the three indexes of the two schemes are nearly the same, namely, the topology management method provided by the embodiment of the invention can provide the same performance as the existing topology management method with indefinite time overhead under most conditions.

Specifically, referring to fig. 8, fig. 8 is a diagram illustrating a comparison of time overhead when the integrated topology management method in the self-organizing network in flight provided by the embodiment of the present invention as shown in fig. 4 is applied to topology management by using the integrated topology management method in the prior art.

In fig. 8, a flying ad-hoc network scene including four task drones moving 10000 time units according to a random walk model, six relay drones, and a ground control station is considered, and in order to adapt to the movement of the task drones, the position of the relay drones needs to be adjusted in time by an algorithm. The graphical time overhead refers to the execution time of both algorithms. As can be seen from FIG. 8, the time overhead of the IFTM scheme is significantly less than that of the PSO-Only scheme. That is to say, the topology management method in the self-organizing network in flight provided by the embodiment of the invention can significantly reduce the time overhead of the topology management process.

According to the topology construction, adjustment and management method in the self-organizing flying network provided by the embodiment of the invention, firstly, in a self-organizing flying network scene, a proper routing protocol and constraint conditions are considered, and a topology construction method based on a particle swarm optimization algorithm is used for deploying and moving relay nodes to complete the construction from zero of an initial topology; then, performing incremental adjustment on the network topology by using a quasi-Newton method to adapt to the random movement of the task unmanned aerial vehicle; and finally, integrating a topology construction method and an adjustment method, managing the topology by using a topology management method, and coping with the accumulated topology change of the flying self-organizing network caused by the random movement of the unmanned aerial vehicle. That is to say, in the technical solution provided in the embodiment of the present invention, a topology construction method based on a particle swarm optimization algorithm is used to complete the zero construction of an initial topology, then a quasi-newton method with low time complexity is used to adjust the topology, and finally a proper threshold is set to measure the degree of change of the topology, thereby implementing comprehensive topology management. Unlike the prior art, the construction from zero of the initial topology is not considered, or the time overhead is not considered in the topology management process. Therefore, compared with the prior art, the topology construction, adjustment and management system in the flying self-organizing network provided by the embodiment of the invention can achieve the purposes of adapting to frequent and rapid topology fluctuation caused by high maneuverability of the unmanned aerial vehicle, dynamically managing the flying self-organizing network topology and reducing the execution time overhead of the topology management algorithm under the condition of maintaining the overall performance of the network, and is simple and convenient to implement, convenient to popularize and wide in application range.

The following provides a description of a system for determining a network topology of a flying ad hoc network according to an embodiment of the present invention.

Referring to fig. 9, fig. 9 is a schematic diagram of a first structure of a topology determining system for an ad hoc network in flight according to an embodiment of the present invention. The system for determining the network topology of the flying ad hoc network provided by the embodiment of the invention comprises the following modules:

a first obtaining module 61, configured to obtain a security constraint condition and a communication constraint condition between end to end of each node in the unmanned aerial vehicle system; the safety constraint condition is that each node in each routing path is in an effective communication range, and the communication constraint condition is that the distance between each node is greater than or equal to the minimum safety distance;

a second obtaining module 62, configured to obtain a routing mechanism of a routing path between each unmanned aerial vehicle and a ground control station corresponding to each unmanned aerial vehicle;

a first processing module 63, configured to determine, based on a longest link distance between links of all routing paths in the routing mechanism, a metric function for evaluating network performance of a flying ad hoc network; the self-organizing network is formed for each unmanned aerial vehicle in the unmanned aerial vehicle system according to the position, the number and the task of each unmanned aerial vehicle in the unmanned aerial vehicle system and the routing mechanism;

a building module 64, configured to build an optimization problem of the network topology of the flying ad hoc network through the security constraint condition, the communication constraint condition and the performance metric function;

and the second processing module 65 is configured to solve the optimization problem of the network topology of the flight ad hoc network through a particle swarm optimization algorithm based on the current position of the relay unmanned aerial vehicle, the current position of the task unmanned aerial vehicle and the flight data of the unmanned aerial vehicle, so as to obtain an initial optimal deployment position of the relay unmanned aerial vehicle, so as to complete optimization of the current position of the relay unmanned aerial vehicle, and determine the initial network topology of the flight ad hoc network.

In a possible implementation manner, the second obtaining module is configured to: acquiring the current positions of all unmanned aerial vehicles and ground control stations, and acquiring a routing mechanism of a routing path between each unmanned aerial vehicle and the ground control station corresponding to each unmanned aerial vehicle by adopting the following protocol expression; wherein, the protocol expression is as follows:

in a possible implementation manner, the first processing module 63 is configured to: based on the longest link distance between links of all routing paths in the routing mechanism, an expression of the following metric function is adopted:

a metric function is determined that evaluates network performance of the flying ad hoc network.

In a possible implementation manner, the building module 64 is configured to build an optimization problem of the flying ad hoc network topology by using the following formula according to the safety constraint, the communication constraint and the performance metric function:

wherein minimize is the three-dimensional position coordinate x of the relay unmanned aerial vehicle by changing_RS is the three-dimensional deployment space of the unmanned aerial vehicle,

for a preset penalty performance indicator function:

in one possible implementation, the second processing module 65 is configured to:

taking the relay unmanned aerial vehicle as particles, wherein all the particles form a population, and the population is distributed in a D-dimensional target search space; wherein, the D dimension is that the position and the flying speed of any particle correspond to a D dimension vector;

after initializing the speed and the position of each particle aiming at each particle, acquiring the optimal position of each particle and the optimal position of a global particle in each iteration;

and iterating all the particles in the target search space according to a speed updating formula and a position updating formula, and updating the position and the speed of each particle until an iteration termination condition is met to obtain the optimal global position of the particles, wherein the optimal global position is used as the initial optimal deployment position of the relay unmanned aerial vehicle.

In one possible implementation, the system further includes:

acquiring the position of each unmanned aerial vehicle, the position of a ground control station, the movement increment of the unmanned aerial vehicle and a routing mechanism in the initial network topology of the self-organizing flying network at the current time step;

and calculating the gradient of the position at each current time step by adopting a quasi-Newton method based on the position of each unmanned aerial vehicle, the position of a ground control station, the movement increment of the unmanned aerial vehicle and a routing mechanism in the initial network topology of the self-organizing flying network at the current time step, and iterating to obtain the optimal adjustment position of each relay unmanned aerial vehicle at the next time step.

In one possible implementation, the system further includes:

the calculation module is used for calculating the topology edit distance of the initial network topology of the self-organizing flying network between the current time step t and the next time step tau so as to measure the change degree of the initial network topology of the self-organizing flying network from the current time step t to the next time step tau;

the second processing module is used for executing the positions of all unmanned aerial vehicles, the positions of ground control stations, the movement increment of the unmanned aerial vehicles and a routing mechanism in the initial network topology of the self-organizing flying network based on the current time step if the topology editing distance is smaller than a first threshold, and iterating to obtain the optimal adjustment position of each relay unmanned aerial vehicle at the next time step by adopting a quasi-Newton method; and if the topology edit distance is larger than a first threshold, executing the updating of the routing path or the reconstruction of the topology so as to reduce the cumulative change of the flying self-organizing network caused by the random movement of each unmanned aerial vehicle.

In one possible implementation manner, the calculation module is configured to:

representing the initial network topology of the flying self-organizing network as a graph when the current time step is long; wherein the graph has a set of nodes, a set of edges, and a set of node locations; the node set is a set of all nodes in the flying ad hoc network, the edge set is a set of all unordered node pairs with the length not greater than a preset length, the node position set is positions of all nodes in the node set, and the nodes comprise: one or more of each unmanned aerial vehicle and a ground control station;

calculating the topology edit distance of the initial network topology of the flying self-organizing network between the current time step and the next time step tau by adopting the following formula based on the graph:

it is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the system embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and reference may be made to the partial description of the method embodiment for relevant points.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for determining network topology of a flying ad hoc network, the method comprising:

the determining a metric function for evaluating network performance of the self-organizing network in flight based on the longest link distance between links of all routing paths in the routing mechanism comprises:

determining a measurement function for evaluating the network performance of the flying self-organizing network;

wherein f is a metric function, x_GIs a three-dimensional position representation of the ground control station, G is a node set of the ground control station, x_MFor three-dimensional position representation of the mission drone, M is a set of nodes, x, of the mission drone_RRepresenting the three-dimensional position of the relay unmanned aerial vehicle, wherein R is a node set of the relay unmanned aerial vehicle, ρ is a routing path between each task unmanned aerial vehicle and a ground control station corresponding to each task unmanned aerial vehicle, k is a node serial number of the routing path between the task unmanned aerial vehicle and the ground control station, and the value range of k is a positive integer,

represents the k task unmanned plane m as a set of ordered nodes_kAnd task unmanned plane m_kCorresponding ground control station g^(m)The m is a task unmanned aerial vehicle, a group of ordered nodes consists of all unmanned aerial vehicles and a ground control station, and the ordered nodes are integrated

In (1), the k-th element in the routing path is represented as

And are assembled

First element of (1)

Must be m_kLast element

Must be that

The remaining elements are relay drones, denoted as

Wherein

For mission unmanned plane m_kAnd a q-th node of the routing path between the ground control station,

for task unmanned plane m_kAnd the q +1 th node of the routing path between the ground control station, | is the cardinal number of the set, δ is the geometric distance between the nodes, and α represents that the link distance influences the task unmanned aerial vehicle m_kAnd the ground control station g^(m)The degree of communication quality;

based on the current position of the relay unmanned aerial vehicle, the current position of the task unmanned aerial vehicle and the flight data of the unmanned aerial vehicle, the optimization problem of the flight ad hoc network topology is solved through a particle swarm optimization algorithm, the initial optimal deployment position of the relay unmanned aerial vehicle is obtained, the current position of the relay unmanned aerial vehicle is optimized, and the initial network topology of the flight ad hoc network is determined.

2. The method of claim 1, wherein the obtaining routing mechanisms for routing paths between each drone and a ground control station corresponding to each drone comprises:

acquiring the current positions of all unmanned aerial vehicles and ground control stations, and acquiring a routing mechanism of a routing path between each unmanned aerial vehicle and the ground control station corresponding to each unmanned aerial vehicle by adopting the following protocol expression; wherein, the protocol expression is as follows:

for the symbolic representation of the mapping relationship,

represents the kth task unmanned aerial vehicle m as a set of ordered nodes_kAnd task unmanned plane m_kCorresponding ground control station g^(m)Route path between, k is a taskThe node sequence number of the routing path between the unmanned aerial vehicle and the ground control station is a task unmanned aerial vehicle, and a group of ordered nodes are composed of the unmanned aerial vehicles and the ground control station and are integrated

In (1), the kth element in the routing path is denoted as

k is a positive integer and is aggregated

First element of (1)

Must be m_kLast element

Must be that

The remaining elements are relay drones, denoted as

Wherein

|. | is the cardinality of the set.

3. The method of claim 1, wherein the optimization problem of the flying ad hoc network topology is constructed by the security constraints, the communication constraints and the performance metric function using the following formula:

wherein minimize is the three-dimensional position coordinate x of the relay unmanned aerial vehicle by changing_RThe minimum penalty performance index function is a three-dimensional deployment space of the unmanned aerial vehicle,

for a preset penalty performance index function:

for task unmanned plane m_kAnd the q +1 th node, d, of the routing path between the ground control station_cmFor effectively communicating unmanned aerial vehiclesThe distance threshold of the message, only when the distance between the nodes is less than the value, the sender and the receiver can keep good communication, cm is the lower corner mark of the distance threshold of the effective communication between the unmanned planes [ ·]⁺Max {0, · }, the physical meaning is to choose the larger one of them, μ is the penalty parameter of the security constraint, and d_sfIn order to avoid the minimum safe distance of collision between unmanned aerial vehicles, the safe distance is far smaller than the end-to-end communication distance, sf is a lower subscript of a minimum safe distance symbol capable of avoiding collision between unmanned aerial vehicles, u is a node u, v is a node v, { lambda { lambda }_m}_m∈MA penalty parameter that is a communication constraint.

4. The method of claim 1, wherein the solving the optimization problem of the flying ad hoc network topology through a particle swarm optimization algorithm to obtain an initial optimal deployment position of the relay drone comprises:

5. The method of claim 1, wherein the method further comprises:

based on the positions of all unmanned aerial vehicles, the positions of the ground control stations, the movement increment of the unmanned aerial vehicles and a routing mechanism in the initial network topology of the self-organizing flying network at the current time step, calculating the gradient of the position at each current time step by adopting a quasi-Newton method, and iterating to obtain the optimal adjustment position of each relay unmanned aerial vehicle at the next time step.

6. The method of claim 1, wherein the method further comprises:

if the topology editing distance is smaller than a first threshold value, executing the positions of all unmanned aerial vehicles, the positions of ground control stations, the movement increment of the unmanned aerial vehicles and a routing mechanism in the initial network topology of the flying ad hoc network based on the current time step, and iterating by adopting a quasi-Newton method to obtain the optimal adjustment position of each relay unmanned aerial vehicle at the next time step;

7. The method of claim 6, wherein calculating a topology edit distance of the initial network topology of the self-organizing flying network between a current time step t and a next time step τ to measure a degree of change of the initial network topology of the self-organizing flying network from the current time step t to the next time step τ comprises:

representing the initial network topology of the flying self-organizing network as a graph when the current time step is long; wherein the graph has a set of nodes, a set of edges, and a set of node locations; the node set is a set of all nodes in the flying ad hoc network, the edge set is a set of all unordered node pairs with lengths not greater than a preset length, the node position set is positions of all nodes in the node set, and the nodes include: one or more of each unmanned aerial vehicle and a ground control station;

wherein, delta_tedFor the topology edit distance, ted is the lower corner mark of the topology edit distance, T (T) is the flying self-organizing network topology at the current time step T, T (T) is the flying self-organizing network topology at the next time step T, and the page_mIs the set of routing paths, w_iWeight parameter for topology editing operation, e_iFor the topology editing operation, t is the current time step, and τ is the next time step.

8. A flying ad hoc network topology determination system, the system comprising:

a first processing module, configured to determine, based on a longest link distance between links of all routing paths in the routing mechanism, a metric function for evaluating network performance of a flying ad hoc network; the self-organizing network is formed by unmanned aerial vehicles in the unmanned aerial vehicle system according to the positions, the number and the tasks of the unmanned aerial vehicles in the unmanned aerial vehicle system and the routing mechanism, and the self-organizing network is based on the routingDetermining a metric function for evaluating network performance of the ad hoc in flight network based on the longest link distance between links of all routing paths in the mechanism comprises: based on the longest link distance between links of all routing paths in the routing mechanism, an expression of the following metric function is employed:

determining a measurement function for evaluating the network performance of the flying self-organizing network; wherein f is a metric function, x_GIs a three-dimensional position representation of the ground control station, G is a node set of the ground control station, x_MIs a three-dimensional position representation of the task UAV, M is a node set of the task UAV, x_RRepresenting the three-dimensional position of the relay unmanned aerial vehicle, wherein R is a node set of the relay unmanned aerial vehicle, rho is a routing path between each task unmanned aerial vehicle and a ground control station corresponding to each task unmanned aerial vehicle, k is a node serial number of the routing path between each task unmanned aerial vehicle and the ground control station, the value range of k is a positive integer,

In (1), the k-th element in the routing path is represented as

And are assembled

First element of (1)

Must be thatm_kLast element

Must be that

The remaining elements are relay drones, denoted as

Wherein

For mission unmanned plane m_kAnd the qth node of the routing path between the ground control stations,

9. The system of claim 8, wherein the second obtaining module is to: acquiring the current positions of all unmanned aerial vehicles and ground control stations, and acquiring a routing mechanism of routing paths between all unmanned aerial vehicles and the ground control stations corresponding to all unmanned aerial vehicles by adopting the following protocol expression; wherein, the protocol expression is as follows:

is a symbolic representation of the mapping relationship,

represents the k task unmanned plane m as a set of ordered nodes_kAnd task unmanned plane m_kCorresponding ground control station g^(m)The route path between the unmanned aerial vehicle and the ground control station, k is the node sequence number of the route path between the unmanned aerial vehicle and the ground control station, m is the unmanned aerial vehicle, and a group of ordered nodes are composed of the unmanned aerial vehicles and the ground control station and are integrated

In (1), the k-th element in the routing path is represented as

k is a positive integer and is aggregated

First element of (1)

Must be m_kLast element

Must be that

The remaining elements are relay drones, denoted as

Wherein

|. | is the cardinality of the set.