CN114205778B - Heterogeneous task-oriented unmanned aerial vehicle cluster cooperative target selection method - Google Patents

Heterogeneous task-oriented unmanned aerial vehicle cluster cooperative target selection method Download PDF

Info

Publication number
CN114205778B
CN114205778B CN202111346648.0A CN202111346648A CN114205778B CN 114205778 B CN114205778 B CN 114205778B CN 202111346648 A CN202111346648 A CN 202111346648A CN 114205778 B CN114205778 B CN 114205778B
Authority
CN
China
Prior art keywords
task
unmanned aerial
aerial vehicle
target
communication
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202111346648.0A
Other languages
Chinese (zh)
Other versions
CN114205778A (en
Inventor
姚昌华
安蕾
韩贵真
高泽郃
程康
胡程程
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Information Science and Technology
Original Assignee
Nanjing University of Information Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing University of Information Science and Technology filed Critical Nanjing University of Information Science and Technology
Priority to CN202111346648.0A priority Critical patent/CN114205778B/en
Publication of CN114205778A publication Critical patent/CN114205778A/en
Application granted granted Critical
Publication of CN114205778B publication Critical patent/CN114205778B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/04—Wireless resource allocation
    • H04W72/044—Wireless resource allocation based on the type of the allocated resource
    • H04W72/0473—Wireless resource allocation based on the type of the allocated resource the resource being transmission power
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/50—Allocation or scheduling criteria for wireless resources
    • H04W72/53—Allocation or scheduling criteria for wireless resources based on regulatory allocation policies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W84/00—Network topologies
    • H04W84/02—Hierarchically pre-organised networks, e.g. paging networks, cellular networks, WLAN [Wireless Local Area Network] or WLL [Wireless Local Loop]
    • H04W84/04—Large scale networks; Deep hierarchical networks
    • H04W84/08—Trunked mobile radio systems
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

The invention discloses an unmanned aerial vehicle cluster cooperative target selection method facing heterogeneous tasks, which constructs a Steinberg game model by considering task values and requirements of different targets and multi-machine cooperative gain and restriction relation, establishes an upper-layer unmanned aerial vehicle as a game leader and a lower-layer unmanned aerial vehicle as a game follower, and provides a distributed strategy updating iterative algorithm to effectively improve the efficiency of the unmanned aerial vehicle cluster system for simultaneously completing a plurality of tasks and realize efficient cooperation facing heterogeneous task values in different environments Neglect the difference of task value, and adopt the not enough, high scheduling of flexibility, computational complexity of centralized scheduling problem.

Description

Heterogeneous task-oriented unmanned aerial vehicle cluster cooperative target selection method
Technical Field
The invention relates to an unmanned aerial vehicle cluster system intelligent optimization technology, in particular to an unmanned aerial vehicle cluster cooperative target selection method for heterogeneous tasks.
Background
In recent years, along with the rapid development of the technical level of artificial intelligence, the intelligent level of unmanned aerial vehicles is higher and higher, a large number of unmanned aerial vehicles form an unmanned aerial vehicle cluster and are applied to various fields of social life, and the unmanned aerial vehicle cluster has wider and wider application potential with high flexibility, wide adaptability and controllable economy, and is highly concerned at home and abroad. The unmanned aerial vehicle is simple, flexible and reliable in equipment, so that selective and targeted observation and communication can be carried out on ground targets in a close range.
The unmanned aerial vehicle cluster system has the advantages of being strong in fault tolerance, good in self-adaptability and the like, and is more suitable for executing tasks in a complex environment. The unmanned aerial vehicle cluster cooperatively completes tasks, which is an important trend in development. In order to improve the benefit of the unmanned aerial vehicle cluster for executing tasks, efficient task allocation must be performed on the unmanned aerial vehicle cluster, and the method is one of key technologies for cooperative control of unmanned aerial vehicles. The field of task planning of unmanned aerial vehicle clusters refers to comprehensive scheduling of target tasks according to requirements of task demands, self characteristics and the like, and therefore a reasonable mapping cooperative relationship between unmanned aerial vehicles and the tasks is established. Although the prior art has partial research on task allocation of multiple drones, most of the research does not consider heterogeneous task values, and most of the considered tasks are isomorphic, and does not consider simultaneous existence of multipoint reconnaissance tasks and communication services. From the aspect of the method, most of the technologies are centralized distribution algorithms, that is, a central control entity is needed to distribute tasks for all members in the cluster, and this mode is not beneficial to improving the robustness and the environment corresponding capability of the unmanned aerial vehicle cluster. Therefore, it is necessary to research a cluster cooperative target selection technology for a cluster of unmanned aerial vehicles to face multiple modes and multiple task values in the process of executing an actual task.
Disclosure of Invention
The purpose is as follows: in order to overcome the defects in the prior art, the invention provides the heterogeneous task-oriented unmanned aerial vehicle cluster cooperative target selection method, so that an unmanned aerial vehicle can reasonably distribute the task objects of each unmanned aerial vehicle in a distributed decision mode through cluster internal cooperation and algorithm iteration according to the requirements and value attributes of target tasks in regions, the overall task capacity of an unmanned aerial vehicle cluster is improved, and the problems that the task distribution of the current unmanned aerial vehicle cluster system is limited to isomorphic tasks and the task values are ignored, the centralized scheduling is insufficient in flexibility, the calculation complexity is high and the like are solved.
The technical scheme is as follows: in order to achieve the purpose, the invention provides the following technical scheme: an unmanned aerial vehicle cluster cooperative target selection method facing heterogeneous tasks is characterized by comprising the following steps:
step 1, initializing the transmitting power and task scheduling of a leading unmanned aerial vehicle, collecting related channel state information and task value of a schedulable task target, and setting an initial running round number and a maximum round number upper limit; let 0 denote the leading drone number, the set of cooperating drones distributed around is denoted as a m ＝[1,2,…,N]The schedulable communication task target set of the leading drone and the cooperating drone is denoted as UE ═ 0,1,2, …, m]The schedulable scout task object set is represented by OE ═ 0,1,2, …, n]Channel gain is g i,j ，j∈UE i ∪OE i ∪{0}，i∈A，A＝A m ∪{0}；
Step 2, firstly, updating and adjusting the strategy of the cooperative unmanned aerial vehicle in the lower-layer sub game, and calculating the utility value corresponding to the selected task target according to the communication reconnaissance task target iteratively selected by each cooperative unmanned aerial vehicle and in combination with the environmental requirement;
the lower level sub-game is defined as:
Figure BDA0003354390060000021
wherein, game participant A m For a set of cooperative drones, a set of policies of participants { Φ } i }，Φ i ＝{p i ,c i }，p i Transmitting power for serving target tasks for each drone, c i For each drone's target task selection, { U i Selecting a utility function value of a target task for each unmanned aerial vehicle;
step 3, repeating the step 2 for iteration to preset times, reasonably distributing the lower-layer task target scheduling, and outputting an optimal strategy set of the lower-layer sub game; the iteration of the lower-layer sub game is updated when each cooperative unmanned aerial vehicle carries out k iterations
Figure BDA0003354390060000022
And then, the sub game is stable, and for any cooperative unmanned aerial vehicle, the target task at the (k +1) th time is selected to be c i (k +1) and kth target task selection c i (k) The difference of the utility values is smaller than a fixed constant zeta, and the optimal strategy set of the N cooperative unmanned aerial vehicles of the lower-layer sub game is output to phi m Wherein phi m ＝{Φ 1 ,Φ 2 ,…,Φ N }。
Step 4, updating and adjusting the strategy of the leading unmanned aerial vehicle in the upper layer game according to the strategy of the lower layer sub-game, and calculating the utility value corresponding to the task of the allocation target according to the communication reconnaissance task target selected by the leading unmanned aerial vehicle;
the upper level sub game is defined as: g ═ A 0 ，{Φ 0 },{U 0 }}
Game participant A 0 Set of policies for participants { Φ, set of leading drones 0 }，Φ i ＝{p 0 ,c 0 }，p 0 Transmitting power to serve target tasks for leading drone, c 0 For the target task selection of the leading drone, { U 0 And selecting a utility function value of the target task for the leading unmanned aerial vehicle.
Step 5, repeating the step 4 for a preset number of times, reasonably distributing the upper layer task target scheduling, and outputting an optimal strategy set of the upper layer sub game; when the leading unmanned aerial vehicle carries out k iterations to satisfy
Figure BDA0003354390060000023
And then, the sub game is stable, and for the leading unmanned aerial vehicle, the task target selection c for the (k +1) th time is performed 0 (k +1) and kth task target selection c 0 (k) The difference of the utility values is smaller than a fixed constant zeta, the upper layer task target scheduling is reasonably distributed, and the optimal strategy set phi of the upper layer sub game leader unmanned aerial vehicle is output 0
And 6, repeating the steps 2-5, iteratively updating the optimal strategy of the upper and lower layer sub-games, and solving and constructing the Steinberg game equilibrium solution
Figure BDA0003354390060000031
Reasonably distributing target tasks of the leading unmanned aerial vehicle and the cooperative unmanned aerial vehicle;
Figure BDA0003354390060000032
the best corresponding strategy for representing the upper game maximization utility function,
Figure BDA0003354390060000033
indicating the best response strategy for the underlying game.
In a preferred embodiment of the present invention, in step 1, the channel gain g is i,j And the task scheduling and power adjusting period is stable and unchanged.
In step 6, a preferred embodiment of the present invention is to combine the strategies
Figure BDA0003354390060000034
The following conditions are satisfied:
Figure BDA0003354390060000035
Figure BDA0003354390060000036
wherein the lower layer game strategy phi m ＝{Φ 1 ,Φ 2 ,…,Φ N }，Φ -i ＝{Φ 0 ,Φ 1 ,…,Φ i-1 ,Φ i+1 ,…,Φ N Represents the strategy combination of the upper-layer leading unmanned aerial vehicle and other cooperative unmanned aerial vehicles at the lower layer, and the optimal strategy of the upper-layer leading unmanned aerial vehicle
Figure BDA0003354390060000037
Given by a lower layer game optimal response strategy, solving by maximizing a self utility function, and obtaining the optimal strategy of each cooperative unmanned aerial vehicle of the lower layer
Figure BDA0003354390060000038
And (4) solving the maximum self utility function by the optimal corresponding strategy of the given upper-layer game and the optimal corresponding strategies of other cooperative unmanned aerial vehicles.
Compared with the prior art, the invention has the following beneficial effects: the invention provides an unmanned aerial vehicle cluster cooperative target selection method facing heterogeneous tasks, which enables unmanned aerial vehicles to reasonably distribute task objects of each unmanned aerial vehicle in a distributed decision mode through cluster internal cooperation and algorithm iteration according to the requirements and value attributes of target tasks in regions, and improves the overall task capacity of an unmanned aerial vehicle cluster.
Meanwhile, the heterogeneous characteristics and the value characteristics of a plurality of tasks in the area are considered, starting from the improvement of the overall task capacity of the cluster, the unmanned aerial vehicle can reasonably distribute the task objects of each unmanned aerial vehicle in a distributed decision-making mode according to the requirements and the value attributes of target tasks in the area through the utility function design and the iterative algorithm design, and the overall task capacity of the unmanned aerial vehicle cluster is improved.
Drawings
Fig. 1 is a diagram of a model of a scout mission distribution system based on unmanned aerial vehicle communication according to an embodiment;
FIG. 2 is a flow chart of a Steinberg equalization-based solution algorithm according to an exemplary embodiment;
FIG. 3 is a diagram of a simulation scenario according to an embodiment;
FIG. 4 is a diagram of convergence of utility functions of a leading unmanned aerial vehicle and a cooperative unmanned aerial vehicle according to an embodiment;
fig. 5 is a target task allocation diagram of a leading drone and a cooperative drone according to an embodiment;
FIG. 6 is a network utility comparison graph according to an exemplary embodiment;
fig. 7 is a graph comparing the utility of networks of different numbers of cooperative drones according to an embodiment.
Detailed Description
The present invention is further illustrated in the accompanying drawings and described in the following detailed description, it is to be understood that such examples are included solely for the purposes of illustration and are not intended as a definition of the limits of the invention, since various equivalent modifications of the invention will become apparent to those skilled in the art after reading the present specification, and it is intended to cover all such modifications as fall within the scope of the invention as defined in the appended claims.
Example (b):
fig. 1 exemplarily shows a scout mission distribution system model based on drone communication, including a leading drone, a cooperating drone and a target. Based on this model, this embodiment provides a heterogeneous task-oriented method for selecting cooperative targets of an unmanned aerial vehicle cluster, and the flow of the method is shown in fig. 2, and includes the following steps:
step 1, initializing the transmitting power and task scheduling of the leading unmanned aerial vehicle, collecting the related channel state information and task value of a schedulable task target, and setting the initial running wheel number and the maximum wheel number upper limit.
Specifically, when 0 denotes the leading drone number, the set of cooperative drones distributed around is denoted as a m ＝[1，2，…N]N is the total number of cooperative drones, and the schedulable communication task target set of the leading drone and the cooperative drones is denoted as UE ═ 0,1,2, … m]M is the total number of schedulable communication task targets, and the set of schedulable scout task targets is represented by OE ═ 0,1,2, … n]And n is the total number of schedulable scout mission targets. Assuming that the channel gain is stable and unchanged in the task scheduling and power adjustment periods, the channel gain of the leader unmanned aerial vehicle, the cooperative unmanned aerial vehicle thereof and the target task is recorded as g i,j Wherein i is equal to A, A is equal to A m Where u {0}, i may refer to a leading drone or a cooperating drone, j ∈ UE ueu {0}, j may refer to a communication and reconnaissance mission target or a leading drone.
In step 2, strategy updating adjustment of the cooperative unmanned aerial vehicles in the lower-layer sub game is firstly carried out, communication reconnaissance task targets iteratively selected by the cooperative unmanned aerial vehicles are combined with environmental requirements, and the utility value corresponding to the selected task targets is calculated.
And the lower layer sub game carries out target task strategy updating and adjusting. In the communication task scheduling, after the scheduling strategies of the leading unmanned aerial vehicle and other cooperative unmanned aerial vehicles are given, the ith cooperative unmanned aerial vehicle CD i Downlink signal-to-noise ratio serving kth communication target task
Figure BDA0003354390060000047
Comprises the following steps:
Figure BDA0003354390060000041
signal-to-noise ratio when uploading communication information
Figure BDA0003354390060000042
Comprises the following steps:
Figure BDA0003354390060000046
wherein the content of the first and second substances,
Figure BDA0003354390060000043
for the ith cooperative unmanned plane CD i The interference sum when the kth communication target task is served comprises cross-layer interference generated by the kth target task served by the leading unmanned aerial vehicle and same-layer interference and noise generated by the kth target task served by other cooperative unmanned aerial vehicles, and p 0 Transmit power for leading drone, g 0,k Channel gain obtained when the leading drone is served the kth target task,
Figure BDA0003354390060000044
generating sum of interference, p, for other cooperative drones -i ＝[p 0 ,p 1 ,…,p i-1 ,p i+1 ,…,p N ]To indicate to remove CD i Power allocation vector, σ, of all but drone 2 Is background interference noise. p is a radical of -i,0 ＝[p 1 ,p 2 ,…p i-1 ,p i+1 ,…,p N ]To indicate to remove CD i In addition to the power allocation vectors of other cooperating drones,
Figure BDA0003354390060000045
indicating CD removal i Sum of interference, p, generated by uploading information by other cooperative unmanned aerial vehicles j,0 Indicating CD removal i Power value g of other cooperative unmanned aerial vehicles uploading information j,0 Indicating CD removal i And other cooperative unmanned aerial vehicles are used for uploading information, so that the channel gain is increased. p is a radical of i,0 Is a CD i Communication upload ofPower vector, assuming CD i The communication uploading rate to LD is R i Allocating a bandwidth of
Figure BDA0003354390060000051
By
Figure BDA0003354390060000052
Can be obtained. For an unmanned aerial vehicle executing a communication service task, the design of the utility function of the unmanned aerial vehicle takes into account the satisfaction degree and the power consumption of a target task at the same time when a CD is given i Serving the kth communication target task, CD i The utility function of (a) can be expressed as:
Figure BDA0003354390060000053
wherein, CD i Communication utility U i k Revenue generation for performing communication tasks for drones
Figure BDA0003354390060000054
And cost consumption
Figure BDA0003354390060000055
The difference and the profit function are partially modeled into an S-shaped function to represent the satisfaction degree of the target task and simultaneously consider the signal-to-noise ratio of downlink communication
Figure BDA0003354390060000056
Satisfaction and communication upload signal-to-noise ratio
Figure BDA0003354390060000057
And theta is a constant and is used for compromising the communication downlink signal-to-noise ratio and the uplink signal-to-noise ratio. In addition, the method can be used for producing a composite material
Figure BDA0003354390060000058
And beta i k For the ith cooperative unmanned plane CD i The steepness and center value of the function for serving the k-th communication target age. val (k) is the value of the kth task object. CD (compact disc) i Cost ofThe function part simultaneously takes into account the power consumption mu for executing the target task i p i Power consumption kappa of uploading communication information i p i,0 And the lower layer CD i Interference penalty λ for upper layer LD communication services i g i,0 p i 。μ i Is a constant, which is used to trade off power consumption, k i Representing the power consumption coefficient, λ, of the uploaded communication information i Representing an interference penalty parameter for adjusting the impact of cross-layer interference on an upper layer service objective task i Increasing the transmission power p i In time, the satisfaction degree of the service target task is increased, and simultaneously higher cross-layer interference is brought to the upper-layer LD, and the Qos of the LD service target task is influenced, so the CD i A compromise optimization is required. For a drone executing a scout service task, the design of the drone scout utility function also includes two parts, namely the satisfaction degree of the target task and the power consumption. In the scout task scheduling, the resolution r of each cooperative unmanned aerial vehicle to each target task is a fixed value, and a resolution matrix is constructed. When given CD i Service scout target task x, CD i The scout utility function of (a) is expressed as:
Figure BDA0003354390060000059
wherein, CD i Communication utility U i x Revenue for performing communication tasks for unmanned aerial vehicles
Figure BDA00033543900600000510
And cost consumption
Figure BDA00033543900600000511
The difference, the gain function, is partially modeled as a sine function, the coefficients are weighted
Figure BDA00033543900600000516
Make the benefit stable to between 0 and 1, r i x Is a CD i For the resolution value of the scout target assignment x,
Figure BDA00033543900600000512
is a CD i Distance from the scout target task x. Wherein the cost function part simultaneously considers the power consumption of the uploading of the scout image
Figure BDA00033543900600000513
Power consumption delta identified by LD i p' i 。p' i For each CD i Total power of scouting tasks, τ i The total power fraction is used to identify the power proportion of the computational process, 1-tau i Represents the power consumption proportion of the upward transmission of the scout information after the identification is finished,
Figure BDA00033543900600000514
interference punishment parameters are uploaded by the scouting information and used for balancing the interference generated by uploading the scouting information on the leading unmanned aerial vehicle i As a constant, balance CD i Power consumption for photographing.
Given the target task selection and transmit power of the upper layer LD, each CD independently selects the best strategy to maximize its utility function, and thus the lower layer sub-game is defined as:
Figure BDA00033543900600000515
the lower layer game G comprises three elements of participants, strategy sets and utility functions, and game participants A m For a set of cooperative drones, a set of policies of participants { Φ } i }，Φ i ＝{p i ,c i }，p i Transmitting power for serving target tasks for each drone, c i And selecting a target task for each unmanned aerial vehicle. { U i And selecting a utility function value of the target task for each unmanned aerial vehicle. Given other drone's policy Φ -i ，CD i Optimal communication target task selection
Figure BDA0003354390060000061
Figure BDA0003354390060000062
Wherein the content of the first and second substances,
Figure BDA0003354390060000063
the interference generated to serve the kth communication target task,
Figure BDA0003354390060000064
for removing CD i Interference generated by uploading information by other cooperative unmanned aerial vehicles, g i,k For cooperating with unmanned aerial vehicle CD i Channel gain, g, obtained while serving the kth target task i,0 Representing cooperative unmanned aerial vehicle CD i Channel gain when uploading information. p is a radical of i,0 ＝ε i p i ，ε i Is a CD i Ratio coefficient of transmission power and uploading power of service communication target task, p i,0 A power vector is uploaded for the communication. Theta is a proportionality coefficient. val (k) is the value of target task k. The CD determines an optimal serving communication objective task and then further optimizes a transmit power maximization communication utility function, which communication utility function is for the CD i Transmission power p i The optimum transmitting power can be obtained by calculating the partial derivative and combining the reciprocal relation of the S-shaped function
Figure BDA0003354390060000065
And utility function
Figure BDA0003354390060000066
Comprises the following steps:
Figure BDA0003354390060000067
Figure BDA0003354390060000068
Figure BDA0003354390060000069
for serving the tth under leading drone and other cooperative drone policies i The interference generated by the individual communication target tasks,
Figure BDA00033543900600000610
for removing CD i Interference generated by uploading information under other cooperative unmanned aerial vehicle strategies,
Figure BDA00033543900600000611
for cooperating with unmanned aerial vehicle CD i T th service i Channel gain, g, obtained at the time of the target task i,0 Representing cooperative unmanned aerial vehicle CD i Channel gain when uploading information. p is a radical of i,0 ＝ε i p i ，ε i Is a CD i And the proportionality coefficient of the transmitting power and the uploading power of the service communication target task. Furthermore alpha i And beta i For the steepness and center value of the sigmoid function,
Figure BDA00033543900600000612
Γ i and theta is the sum of the uplink signal-to-noise ratio and the downlink signal-to-noise ratio of the service target task, and is a proportionality coefficient.
Figure BDA00033543900600000613
Wherein mu i Is a constant, which is used to trade off power consumption, k i Representing the power consumption coefficient, epsilon, of the uploaded communication information i Is a CD i Ratio of transmission power to upload power, lambda, of serving communication target task i Denotes an interference penalty parameter, g i,0 Representing cooperative unmanned aerial vehicle CD i Channel gain in uploading information, val (t) i ) Is a target task t i The value of (A) is obtained. Similarly, let total power p 'of reconnaissance target task' i ＝p i Finding out the effectiveness of the scout network, comparing the task effectiveness values of all the scout targets, and selecting the optimal scout target
Figure BDA00033543900600000614
If the utility value is negative, p' i If the result is equal to 0, the scout mission of the target is selected to be abandonedAnalyzing communication scouting network utility and determining optimal task target selection
Figure BDA00033543900600000615
The service target task iteration strategy under the final strategy updating iteration is as follows:
Figure BDA00033543900600000616
Figure BDA0003354390060000071
for optimal communication target task
Figure BDA0003354390060000072
The utility of the network is that,
Figure BDA0003354390060000073
for optimal scouting of target tasks
Figure BDA0003354390060000074
And the utility is compared with the utility value to adjust the target task selection.
In the step 3, the iteration of the step 2 is repeated to preset times, the lower layer task scheduling is reasonably distributed, and the optimal strategy set of the lower layer sub game is output;
the iteration of the lower-layer sub game is updated when each cooperative unmanned aerial vehicle carries out k iterations
Figure BDA0003354390060000075
After that, the sub-game stabilizes. For any cooperative unmanned aerial vehicle, selecting c for target task at k +1 time i (k +1) and kth target task selection c i (k) The difference of the utility values is smaller than a fixed constant zeta, the lower layer task target tasks are dispatched and reasonably distributed, and the optimal strategy set phi of the lower layer sub game N cooperative unmanned aerial vehicles is output m Wherein phi m ＝{Φ 1 ,Φ 2 ,…,Φ N }。
In step 4, updating and adjusting the strategy of the leading unmanned aerial vehicle in the upper layer game according to the strategy of the lower layer sub-game, and calculating the utility value corresponding to the task of the allocation target according to the communication reconnaissance task target selected by the leading unmanned aerial vehicle;
and the upper layer sub game carries out target task strategy updating and adjusting. In the communication task scheduling, after the scheduling strategy of each collaborative unmanned aerial vehicle CD of the lower layer is given, the downlink signal-to-noise ratio of the I-th user of LD service
Figure BDA0003354390060000076
Can be expressed as:
Figure BDA0003354390060000077
wherein the content of the first and second substances,
Figure BDA0003354390060000078
representing the sum of interference p generated by each lower cooperative unmanned aerial vehicle on LD service target task l 0 For leading the unmanned aerial vehicle transmitting power, g 0,l Serving the channel gain of target task l for the leading drone. p is a radical of j And g j,l The transmit power for other coordinated drones and the gain for target task l. Sigma 2 Is noise. For an unmanned aerial vehicle executing a communication service task, the design of the utility function of the unmanned aerial vehicle takes into account the satisfaction degree and power consumption of a target task, and for a given communication target task k, the utility function of the leading unmanned aerial vehicle LD can be expressed as:
Figure BDA0003354390060000079
the utility function U 0 k Comprises two parts, the first part is the benefit of the service communication target task
Figure BDA00033543900600000710
Modeled as an S-shaped function representing the value of the benefit from the satisfaction of the serving communication objective task, and the second part is a cost function
Figure BDA00033543900600000711
Representing dynamic power overhead, wherein the parameter α 0 And beta 0 Respectively the steepness and the center value of the sigmoid function. val (k) represents the value of the communication task destination LDk,
Figure BDA00033543900600000712
and serving the LD with the downlink signal-to-noise ratio of the target task k. p is a radical of formula 0 Pilot unmanned aerial vehicle transmitting power, mu 0 Is a constant used to balance the satisfaction of the target object of the service task and the power energy consumption. For an unmanned aerial vehicle executing a scout service task, the design of the unmanned aerial vehicle scout utility function also comprises two parts of satisfaction degree and power consumption of a target task, in the scout task scheduling, the resolution r of each cooperative unmanned aerial vehicle or leading unmanned aerial vehicle to each task target is a fixed value, and a resolution matrix is constructed.
Figure BDA0003354390060000081
The utility function U 0 x Comprises two parts which are respectively connected with a power supply and a power supply,
Figure BDA0003354390060000082
representing the revenue of the service reconnaissance mission objective,
Figure BDA0003354390060000083
represents the cost of the service scout mission objective, i.e., the power consumption of LD image recognition, wherein,
Figure BDA00033543900600000821
for the resolution of the LD to the task object x,
Figure BDA0003354390060000084
is the distance between the LD and the task target x,
Figure BDA0003354390060000085
the coefficient is traded off to make the benefit stable with 0-1. Delta 0 Is an image recognition power consumption proportional constant, p' 0 Power value when reconnaissance is carried out for leading unmanned aerial vehicle.
Optimal strategy set phi according to lower layer sub game m And the upper-layer leader unmanned plane LD independently selects the optimal strategy to maximize the utility function of the upper-layer leader unmanned plane LD, so that the upper-layer sub game is defined as follows:
G＝{A 0 ，{Φ 0 },{U 0 }}
similarly, the upper layer game G has three elements of a participant, a strategy set and a utility function, and the game participant A 0 For the set of leading drones, the set of policies of the participants { Φ } 0 }，Φ i ＝{p 0 ,c 0 }，p 0 Transmitting power to serve target tasks for leading drone, c 0 And selecting a target task of the leading unmanned aerial vehicle. { U 0 And selecting a utility function value of the target task for the leading unmanned aerial vehicle. LD optimal communication task target selection given other drone policies
Figure BDA0003354390060000086
Figure BDA0003354390060000087
Wherein the content of the first and second substances,
Figure BDA0003354390060000088
interference generated to serve the kth communication task objective, g 0,k The channel gain value obtained when the kth target task is served for the leading drone LD, val (k) is the value of the target task k. The LD determines the optimal service communication target task, then further optimizes the transmission power to maximize the communication utility function, and combines the reciprocal relation of the S-shaped function to obtain the optimal transmission power
Figure BDA0003354390060000089
And utility function
Figure BDA00033543900600000810
Comprises the following steps:
Figure BDA00033543900600000811
Figure BDA00033543900600000812
Figure BDA00033543900600000813
wherein alpha is 0 And beta 0 For the steepness and central value of the sigmoid function, gamma 0 Downstream signal-to-noise ratio, val (t), for serving a communication target 0 ) For task target t 0 Value of (a), mu 0 Is a constant that balances the satisfaction of the service objective with the power energy consumption. And similarly, solving the network utility of the upper layer scout task target, comparing the utility values of all the scout task targets, and selecting the optimal scout task
Figure BDA00033543900600000814
Analyzing communication scouting network utility and determining optimal task target selection
Figure BDA00033543900600000815
The service task target iteration strategy under the final strategy updating iteration is as follows:
Figure BDA00033543900600000816
Figure BDA00033543900600000817
for optimal communication task goal
Figure BDA00033543900600000818
The utility of the network is that,
Figure BDA00033543900600000819
for optimal scouting mission objectives
Figure BDA00033543900600000820
And the utility is compared with the utility value to adjust the target task selection.
In step 5, repeating the step 4 until the preset times, reasonably distributing the upper layer task target scheduling, and outputting an optimal strategy set of the upper layer sub game;
the iteration of the upper layer sub game is updated, and when the leading unmanned aerial vehicle carries out k iterations
Figure BDA0003354390060000091
After that, the sub-game is stable. For leading unmanned aerial vehicle, target task selection c for k +1 time 0 (k +1) and kth target task selection c 0 (k) The difference of the utility values is smaller than a fixed constant zeta, the upper layer task target scheduling is reasonably distributed, and the optimal strategy set phi of the upper layer sub game leader unmanned aerial vehicle is output 0 。
And 6, repeating the steps 2-5, iteratively updating the optimal strategy of the upper and lower layer sub-games, establishing the balance of the Steinberg game, and reasonably distributing the target tasks of the leading unmanned aerial vehicle and the cooperative unmanned aerial vehicle.
Figure BDA0003354390060000092
Represents the best corresponding strategy of the upper game maximization utility function,
Figure BDA0003354390060000093
indicating the best response strategy for the underlying game. For any combination of strategies, the following conditions are satisfied:
Figure BDA0003354390060000094
Figure BDA0003354390060000095
wherein the lower layer game strategy phi m ＝{Φ 1 ,Φ 2 ,…,Φ N }，Φ -i ＝{Φ 0 ,Φ 1 ,…,Φ i-1 ,Φ i+1 ,…,Φ N Represents the strategy combination of the upper-layer leading unmanned aerial vehicle and the lower-layer other cooperative unmanned aerial vehicles,
Figure BDA0003354390060000096
known as steinberg equalization. Optimal strategy of upper-layer leading unmanned aerial vehicle
Figure BDA0003354390060000097
The method is given by a lower-layer game optimal response strategy, and the maximum self utility function is solved. In a similar way, the optimal strategy of each cooperative unmanned aerial vehicle at the lower layer
Figure BDA0003354390060000098
The optimal corresponding strategy of the upper-layer game and the optimal corresponding strategies of other cooperative unmanned aerial vehicles are given, and the self utility function is maximized to solve.
Fig. 2 shows an algorithm flow chart of the method, the sub-game is solved by a general iterative algorithm in a circulating mode, the Steinberg equilibrium iteration is finished, the upper and lower layer target task distribution is not changed any more, the equilibrium of the upper and lower layer sub-games is sought mainly through a reverse recursion method, and the task distribution problem in the multi-unmanned aerial vehicle system is realized.
As shown in fig. 3, the radius of the LD serviceable task target area is 300m, 15 CDs are randomly distributed in the LD scheduling range, the radius of the serviceable task target is 30m, and the communication and reconnaissance tasks are randomly distributed in the LD and CD service ranges. The LD can serve 3 communication task targets and reconnaissance task targets. The number of communication and reconnaissance task objects that can be served by a CD is 4,5,4,5,4,4,5,3,4,3 and 3,4,2,4,3,2,4,1,2,2 in sequence. The communication and reconnaissance task target value val served by the LD is 1, the communication and reconnaissance task target value served by the CD is relatively low, and the values are [0.9,0.95 ]]And (4) internally generating randomly. Wherein the CD i To task object jChannel gain
Figure BDA0003354390060000099
Representing the corresponding distance, the signal attenuation is 25 dB. The target signal-to-noise ratio of the communication task served by the LD is gamma 0 30dB, the signal-to-noise ratio of the communication task target served by the CD and the uploading signal-to-noise ratio are both 10,20]Generated randomly within dB. Noise power σ 2 ＝10 -8 mW. parameter alpha i 1, and theta is 1. Communication interference penalty and interference cost parameter set to lambda i ＝10 8 ，μ i 1/mV with an upload power consumption parameter of κ i 1/mV. LD identification image power consumption delta in scout task 0 1, parameter of
Figure BDA0003354390060000101
CD recognition image power consumption delta i 1, the upload interference penalty and the upload power ratio parameter are set to
Figure BDA0003354390060000102
τ i 0.6. Weighing parameters
Figure BDA0003354390060000103
The resolution of the LD and CD for the scout mission target object is given in table 1.
TABLE 1 LD and CD pairs scout target resolution
Figure BDA0003354390060000104
Fig. 4 shows a network utility iteration update curve of the corresponding leading drone and the cooperative drone according to the method of the present invention, where 30 rounds are set for each iteration until a preset number of times. From the update curve of network utility, the convergence state can be finally achieved after the upper and lower layers of game interaction iteration, and the convergence performance of the algorithm is verified.
Fig. 5 shows the optimal communication and reconnaissance mission allocation of the unmanned aerial vehicles when the upper and lower sub-games reach the steinberg balance by using the method of the invention, and each unmanned aerial vehicle can independently perform the optimal allocation of service communication or reconnaissance target missions.
Fig. 6 shows the system utility value change in three states of jointly considering communication and scout tasks, only considering communication tasks and only considering scout tasks in the target task scheduling process by using the method of the present invention. The utility values of the system under the condition of 8, 9, 10, 11, 12 and 13 cooperative unmanned aerial vehicles are respectively shown in fig. 7, it can be seen that the upper and lower layer games finally converge to the equilibrium point, and the utility values of the systems of the used algorithms are all larger than the system utility value under the condition of considering communication or reconnaissance of a single index.
The multi-unmanned aerial vehicle communication and reconnaissance task allocation has important research significance in unmanned cluster network optimization. The method focuses on the combined optimization of target task scheduling and power control in the unmanned aerial vehicle network, utilizes a layered game framework to analyze decision behaviors of a leading unmanned aerial vehicle and a cooperative unmanned aerial vehicle, adopts a distributed strategy iterative updating algorithm to realize Stackelberg (Stackelberg) balance, realizes the optimal target task scheduling of the unmanned aerial vehicle, performs simulation analysis on a plurality of scenes, verifies that the provided algorithm can realize the convergence of distributed task allocation and system stability in the multi-unmanned aerial vehicle system, and effectively improves the overall effectiveness of tasks executed by the multi-unmanned aerial vehicle system.
The above description is only of the preferred embodiments of the present invention, and it should be noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the invention and these are intended to be within the scope of the invention.

Claims (2)

1. An unmanned aerial vehicle cluster cooperative target selection method facing heterogeneous tasks is characterized by comprising the following steps:
step 1, initializing the transmitting power and task scheduling of a leading unmanned aerial vehicle, collecting related channel state information and task value of a schedulable task target, and setting an initial running round number and a maximum round number upper limit; let 0 denote the leading drone number, then the set of cooperative drones distributed around is denoted as a m ＝[1,2,…,N]N is the total number of the cooperative unmanned aerial vehiclesHead drone and cooperative drone schedulable communication task target set denoted UE ═ 0,1,2, …, m]M is the total number of schedulable communication task targets, and the set of schedulable scout task targets is represented by OE ═ 0,1,2, …, n]N is the total number of schedulable scout mission targets and the channel gain is g i,j ，j∈UE i ∪OE i ∪{0}，i∈A，A＝A m ∪{0}；
Step 2, firstly, strategy updating adjustment of the cooperative unmanned aerial vehicles in the lower-layer sub game is carried out, communication reconnaissance task targets iteratively selected by the cooperative unmanned aerial vehicles are combined with environmental requirements, and the utility value corresponding to the selected task targets is calculated;
and (3) updating and adjusting the target task strategy by the lower layer sub game: in the communication task scheduling, after the scheduling strategies of the leading unmanned aerial vehicle and other cooperative unmanned aerial vehicles are given, the ith cooperative unmanned aerial vehicle CD i Downlink signal-to-noise ratio serving kth communication target task
Figure FDA0003746644320000011
Comprises the following steps:
Figure FDA0003746644320000012
signal-to-noise ratio when uploading communication information
Figure FDA0003746644320000013
Comprises the following steps:
Figure FDA0003746644320000014
wherein the content of the first and second substances,
Figure FDA0003746644320000015
Figure FDA0003746644320000016
for the ith cooperative unmanned plane CD i Serving the kth communicationThe interference sum during the target task comprises cross-layer interference generated by the kth target task served by the leading unmanned aerial vehicle and same-layer interference and noise generated by the kth target task served by other cooperative unmanned aerial vehicles, and p 0 For the transmission power of the leading drone, g 0,k Channel gain obtained when the leading drone is served the kth target task,
Figure FDA0003746644320000017
generating sum of interference, p, for other cooperative drones -i ＝[p 0 ,p 1 ,…,p i-1 ,p i+1 ,…,p N ]Express to remove CD i Power allocation vectors, σ, of all but unmanned aerial vehicles 2 Is background interference noise; p is a radical of -i,0 ＝[p 1 ,p 2 ,…p i-1 ,p i+1 ,…,p N ]To indicate to remove CD i In addition to the power allocation vectors of other cooperating drones,
Figure FDA0003746644320000018
indicating CD removal i The sum of interference, p, generated by uploading information by other cooperative unmanned aerial vehicles j,0 Indicating CD removal i Power value g of information uploaded by other cooperative unmanned aerial vehicles j,0 Indicating CD removal i Channel gain when other cooperative unmanned aerial vehicles upload information; p is a radical of i,0 Is a CD i Of the communication upload power vector, assuming CD i The LD communication uploading rate of the unmanned aerial vehicle to the leader is R i Allocating a bandwidth of
Figure FDA0003746644320000019
By
Figure FDA00037466443200000110
Obtaining; for a drone performing a communication service task, the design of the utility function of the drone takes into account both the satisfaction of the target task and the power consumption when given a CD i Serving the kth communication target task, CD i The utility function of (a) is expressed as:
Figure FDA0003746644320000021
wherein, CD i Communication utility of
Figure FDA0003746644320000022
Revenue generation for performing communication tasks for drones
Figure FDA0003746644320000023
And cost consumption
Figure FDA0003746644320000024
The difference and the profit function are partially modeled into an S-shaped function to represent the satisfaction degree of the target task and simultaneously consider the signal-to-noise ratio of downlink communication
Figure FDA0003746644320000025
Satisfaction and communication upload signal-to-noise ratio
Figure FDA0003746644320000026
Theta is a constant and is used for compromising the signal-to-noise ratio of the communication downlink and the signal-to-noise ratio of the communication uplink; in addition, alpha i k And beta i k For the ith cooperative unmanned plane CD i Serving the steepness and center value of the function for the k-th communication target age; val (k) is the value of the kth task goal; CD (compact disc) i The cost function part simultaneously considers the power consumption mu of executing the target task i p i Power consumption k for uploading communication information i p i,0 And the lower layer CD i Interference punishment lambda for LD communication service of upper-layer leading unmanned aerial vehicle i g i,0 p i ；μ i Is a constant, which is used to trade off power consumption, k i Representing the power consumption coefficient, λ, of the uploaded communication i Representing an interference penalty parameter for adjusting the effect of cross-layer interference on an upper layer service target task, when the CD is used i Increasing the transmission power p i Clothes for manThe satisfaction degree of the service target task is increased, higher cross-layer interference can be brought to the upper-layer leading unmanned aerial vehicle LD, the Qos of the target task served by the leading unmanned aerial vehicle LD is influenced, and therefore the CD i Compromise optimization is required; for the unmanned aerial vehicle executing the reconnaissance service task, the design of the reconnaissance utility function of the unmanned aerial vehicle also comprises two parts of satisfaction degree and power consumption of a target task; in the scout task scheduling, the resolution r of each cooperative unmanned aerial vehicle to each target task is a fixed value, and a resolution matrix is constructed; when given CD i Service scout target task x, CD i The scout utility function of (a) is expressed as:
Figure FDA0003746644320000027
wherein, CD i Communication utility of
Figure FDA0003746644320000028
Revenue for performing communication tasks for unmanned aerial vehicles
Figure FDA0003746644320000029
And cost consumption
Figure FDA00037466443200000210
The difference, the gain function, is partially modeled as a sine function, the trade-off coefficients
Figure FDA00037466443200000211
Make the benefit stable to between 0 and 1, r i x Is a CD i For the resolution value of the reconnaissance target task x,
Figure FDA00037466443200000212
is a CD i Distance from the reconnaissance target task x, val (x) representing task value; wherein the cost function part simultaneously considers the power consumption of the uploading of the scout image
Figure FDA00037466443200000213
And power consumption delta identified by lead unmanned plane LD i p' i ；p' i For each CD i Total power of scouting tasks, τ i The total power fraction is used to identify the power proportion, 1-tau, of the computational processing i Represents the power consumption proportion of the upward transmission of the scout information after the identification is finished,
Figure FDA00037466443200000214
interference punishment parameters are uploaded to the reconnaissance information and used for balancing interference, delta, generated by the reconnaissance information uploading on the leading unmanned aerial vehicle i As a constant, balance CD i Power consumption for taking a picture;
given the target task selection and the transmitting power of the upper-layer leading unmanned aerial vehicle LD, each CD independently selects the optimal strategy to maximize the utility function of the CD, and the lower-layer sub game is defined as follows:
Figure FDA0003746644320000031
wherein, game participant A m For the set of cooperative drones, the policy set of participants { Φ } i }，Φ i ＝{p i ,c i }，p i Transmitting power for serving target tasks for each drone, c i For each drone's target task selection,
Figure FDA0003746644320000032
selecting a utility function value of the target task for each unmanned aerial vehicle;
given other drone's policy Φ -i ，CD i Optimal communication target task selection
Figure FDA0003746644320000033
Figure FDA0003746644320000034
Wherein the content of the first and second substances,
Figure FDA0003746644320000035
the interference generated to serve the kth communication target task,
Figure FDA0003746644320000036
for removing CD i Interference generated by uploading information by other cooperative unmanned aerial vehicles, g i,k For cooperating with unmanned aerial vehicle CD i Channel gain, g, obtained while serving the kth target task i,0 Representing cooperative unmanned aerial vehicle CD i Channel gain when uploading information; p is a radical of i,0 ＝ε i p i ，ε i Is a CD i Ratio coefficient of transmission power and uploading power of service communication target task, p i,0 Uploading a power vector for the communication; theta is a proportionality coefficient; val (k) is the value of target task k; the CD determines an optimal serving communication objective task and then further optimizes a transmit power maximization communication utility function, which is applied to the CD i Transmission power p i The optimum transmitting power can be obtained by combining the relation of S-shaped function reciprocal
Figure FDA0003746644320000037
And utility function
Figure FDA0003746644320000038
Comprises the following steps:
Figure FDA0003746644320000039
Figure FDA00037466443200000310
Figure FDA00037466443200000311
Figure FDA00037466443200000312
for serving the tth under leading drone and other cooperative drone policies i The interference generated by the individual communication target tasks,
Figure FDA00037466443200000313
for removing CD i Interference generated by uploading information under other cooperative unmanned aerial vehicle strategies,
Figure FDA00037466443200000314
for cooperating with unmanned CD i T th service i Channel gain, g, obtained at the time of the target task i,0 Representing cooperative unmanned aerial vehicle CD i Channel gain when uploading information; p is a radical of i,0 ＝ε i p i ，ε i Is a CD i The ratio coefficient of the transmitting power and the uploading power of the service communication target task; in addition, alpha i And beta i For the steepness and center value of the sigmoid function,
Figure FDA00037466443200000315
Γ i the sum of the uplink signal-to-noise ratio and the downlink signal-to-noise ratio of the service target task is obtained, and theta is a proportionality coefficient;
Figure FDA00037466443200000316
wherein mu i Is a constant, which is used to trade off power consumption, k i Representing the power consumption coefficient, epsilon, of the uploaded communication information i Is a CD i Ratio of transmission power to upload power, lambda, of serving communication target task i Denotes an interference penalty parameter, g i,0 Representing cooperative unmanned aerial vehicle CD i Channel gain in uploading information, val (t) i ) Is a target task t i The value of (D); similarly, let total power p 'of reconnaissance target task' i ＝p i Finding out the effectiveness of the scout network, comparing the task effectiveness values of all the scout targets, and selecting the optimal scout target
Figure FDA0003746644320000041
If the utility value is negative, p' i When the task is equal to 0, selecting the scout task for abandoning the target, analyzing the communication scout network utility, and determining the optimal task target selection
Figure FDA0003746644320000042
The service target task iteration strategy under the final strategy updating iteration is as follows:
Figure FDA0003746644320000043
Figure FDA0003746644320000044
for optimal communication target task
Figure FDA0003746644320000045
The utility of the network is that,
Figure FDA0003746644320000046
for optimal scouting of target tasks
Figure FDA0003746644320000047
Comparing the utility value to adjust the selection of the target task;
step 3, repeating the step 2 for iteration to preset times, reasonably distributing the lower-layer task target scheduling, and outputting an optimal strategy set of the lower-layer sub game; the iteration of the lower-layer sub game is updated when each cooperative unmanned aerial vehicle carries out k iterations
Figure FDA0003746644320000048
And then, the sub game is stable, and for any cooperative unmanned aerial vehicle, the target task at the (k +1) th time is selected to be c i (k +1) and kth target task selection c i (k) The difference of the utility values is smaller than a fixed constant zeta, and the most cooperative unmanned aerial vehicle of the lower-layer sub game N frames is outputSet of optimal strategies as phi m Wherein phi m ＝{Φ 1 ,Φ 2 ,…,Φ N }；
Step 4, updating and adjusting the strategy of the leading unmanned aerial vehicle in the upper layer game according to the strategy of the lower layer sub-game, and calculating the utility value corresponding to the task of the allocation target according to the communication reconnaissance task target selected by the leading unmanned aerial vehicle;
the upper layer sub game carries out target task strategy updating adjustment; in the communication task scheduling, when the scheduling strategy of each collaborative unmanned aerial vehicle CD of the lower layer is given, the downlink signal-to-noise ratio of the I-th user served by the LD of the leading unmanned aerial vehicle
Figure FDA0003746644320000049
Can be expressed as:
Figure FDA00037466443200000410
wherein the content of the first and second substances,
Figure FDA00037466443200000411
representing the sum of interference p generated by each lower cooperative unmanned aerial vehicle on LD service target task l of the leading unmanned aerial vehicle 0 For leading the unmanned aerial vehicle transmitting power, g 0,l Channel gain of a target task l is served for the leading unmanned aerial vehicle; p is a radical of j And g j,l The gain for the transmission power of other cooperative unmanned aerial vehicles and the target task l; sigma 2 Is noise; for an unmanned aerial vehicle executing a communication service task, the design of the utility function of the unmanned aerial vehicle takes into account the satisfaction degree and power consumption of a target task, and for a given communication target task k, the utility function of the leading unmanned aerial vehicle LD can be expressed as:
Figure FDA00037466443200000412
the utility function
Figure FDA00037466443200000413
Comprises two parts, the first part is the benefit of the service communication target task
Figure FDA00037466443200000414
Modeled as an S-shaped function, representing the value of the benefit from the satisfaction of the serving communication objective task, and the second part is a cost function
Figure FDA00037466443200000415
Representing dynamic power overhead, where the parameter α 0 And beta 0 The steepness and the central value of the S-shaped function are respectively; val (k) represents the value of the communication mission target lead drone LD k,
Figure FDA00037466443200000416
serving a downlink signal-to-noise ratio of a target task k for a leading unmanned aerial vehicle LD; p is a radical of 0 Pilot unmanned plane transmitting power, mu 0 The constant is used for balancing the satisfaction degree of the target object of the service task and the power energy consumption; for the unmanned aerial vehicle executing the reconnaissance service task, the design of the reconnaissance utility function of the unmanned aerial vehicle also comprises two parts of satisfaction degree and power consumption of a target task, in the reconnaissance task scheduling, the resolution r of each cooperative unmanned aerial vehicle or the leading unmanned aerial vehicle to each task target is a fixed value, a resolution matrix is constructed, and when a leading unmanned aerial vehicle LD serves a reconnaissance task target x, the reconnaissance utility function of the leading unmanned aerial vehicle LD is expressed as follows:
Figure FDA0003746644320000051
the utility function
Figure FDA0003746644320000052
Comprises two parts which are respectively connected with a power supply and a power supply,
Figure FDA0003746644320000053
receiving of object representing service scout taskThe advantages that the method is good for,
Figure FDA0003746644320000054
represents the cost of serving the scout mission objective, i.e., the power consumption of the leading drone LD image recognition, where,
Figure FDA0003746644320000055
to get the resolution of the drone LD to the task target x,
Figure FDA0003746644320000056
for the distance between the leading unmanned plane LD and the task target x,
Figure FDA0003746644320000057
the coefficient is balanced to ensure that the benefit is stable and is between 0 and 1; delta 0 Identifying Power consumption proportionality constant, p 'for an image' 0 Power value for the leading unmanned aerial vehicle during reconnaissance;
optimal strategy set phi according to lower-layer sub game m The upper-layer leader unmanned aerial vehicle LD independently selects an optimal strategy to maximize a self utility function, and the upper-layer sub game is defined as follows:
Figure FDA0003746644320000058
game participant A 0 Set of policies for participants { Φ, set of leading drones 0 }，Φ i ＝{p 0 ,c 0 }，p 0 Transmitting power to serve target tasks for leading drone, c 0 For the target mission selection of the leading drone,
Figure FDA0003746644320000059
selecting a utility function value of the target task for the leading unmanned aerial vehicle;
Figure FDA00037466443200000510
selecting a utility function value of the target task for the leading unmanned aerial vehicle; given aStrategy of other unmanned aerial vehicles, optimal communication task target selection of leading unmanned aerial vehicle LD
Figure FDA00037466443200000511
Figure FDA00037466443200000512
Wherein the content of the first and second substances,
Figure FDA00037466443200000513
interference generated to serve the kth communication task objective, g 0,k A channel gain value, val (k), obtained when the leading drone LD serves the kth target task is the value of target task k; determining an optimal service communication target task by a leading unmanned aerial vehicle LD, further optimizing a transmission power maximization communication utility function, and obtaining the optimal transmission power by combining the relation of the reciprocal of an S-shaped function
Figure FDA00037466443200000514
And utility function
Figure FDA00037466443200000515
Comprises the following steps:
Figure FDA00037466443200000516
Figure FDA00037466443200000517
Figure FDA0003746644320000061
wherein alpha is 0 And beta 0 For the steepness and central value of the sigmoid function, gamma 0 Downstream signal-to-noise ratio, val (t), for serving a communication target 0 ) To a task objectt 0 Value of (a), mu 0 Is a constant used to trade off satisfaction of service objectives and power energy consumption; and similarly, solving the network utility of the upper layer scout task target, comparing the utility values of all the scout task targets, and selecting the optimal scout task
Figure FDA0003746644320000062
Analyzing communication scouting network utility and determining optimal task target selection
Figure FDA0003746644320000063
The service task target iteration strategy under the final strategy updating iteration is as follows:
Figure FDA0003746644320000064
Figure FDA0003746644320000065
for optimal communication task goal
Figure FDA0003746644320000066
The utility of the network is that,
Figure FDA0003746644320000067
for optimal scout mission objectives
Figure FDA0003746644320000068
Comparing the utility value to adjust the selection of the target task;
step 5, repeating the step 4 for a preset number of times, reasonably distributing the upper layer task target scheduling, and outputting an optimal strategy set of the upper layer sub game; when the leading unmanned aerial vehicle carries out k iterations to satisfy
Figure FDA0003746644320000069
Later, the sub-game is stable, for leading unmanned, kth+1 task target selection c 0 (k +1) and kth task object selection c 0 (k) The difference of the utility values is smaller than a fixed constant zeta, the upper layer task target scheduling is reasonably distributed, and the optimal strategy set phi of the upper layer sub game leader unmanned aerial vehicle is output 0 ；
And 6, repeating the steps 2-5, iteratively updating the optimal strategy of the upper and lower layer sub-games, and solving and constructing the Steinberg game equilibrium solution
Figure FDA00037466443200000610
Reasonably distributing target tasks of the leading unmanned aerial vehicle and the cooperative unmanned aerial vehicle;
Figure FDA00037466443200000611
the best corresponding strategy for representing the upper game maximization utility function,
Figure FDA00037466443200000612
representing the best response strategy for the underlying game, combining the strategies
Figure FDA00037466443200000613
The following conditions are satisfied:
Figure FDA00037466443200000614
Figure FDA00037466443200000615
wherein the lower layer game strategy phi m ＝{Φ 1 ,Φ 2 ,…,Φ N }，Φ -i ＝{Φ 0 ,Φ 1 ,…,Φ i-1 ,Φ i+1 ,…,Φ N Represents the strategy combination of the upper-layer leading unmanned aerial vehicle and the lower-layer other cooperative unmanned aerial vehicles,
Figure FDA00037466443200000616
optimal strategy called Stenberg equilibrium, top leader drone
Figure FDA00037466443200000617
Given by the lower-layer game optimal response strategy, solving by the maximized self utility function, and determining the optimal strategy of each cooperative unmanned aerial vehicle in the lower layer
Figure FDA00037466443200000618
And (4) solving the maximum self utility function by the optimal corresponding strategy of the given upper-layer game and the optimal corresponding strategies of other cooperative unmanned aerial vehicles.
2. The heterogeneous task-oriented unmanned aerial vehicle cluster cooperative target selection method according to claim 1, wherein in step 1, the channel gain g is i,j And the task scheduling and power adjusting period is stable and unchanged.
CN202111346648.0A 2021-11-15 2021-11-15 Heterogeneous task-oriented unmanned aerial vehicle cluster cooperative target selection method Active CN114205778B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111346648.0A CN114205778B (en) 2021-11-15 2021-11-15 Heterogeneous task-oriented unmanned aerial vehicle cluster cooperative target selection method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111346648.0A CN114205778B (en) 2021-11-15 2021-11-15 Heterogeneous task-oriented unmanned aerial vehicle cluster cooperative target selection method

Publications (2)

Publication Number Publication Date
CN114205778A CN114205778A (en) 2022-03-18
CN114205778B true CN114205778B (en) 2022-08-26

Family

ID=80647441

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111346648.0A Active CN114205778B (en) 2021-11-15 2021-11-15 Heterogeneous task-oriented unmanned aerial vehicle cluster cooperative target selection method

Country Status (1)

Country Link
CN (1) CN114205778B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108632831A (en) * 2018-05-11 2018-10-09 南京航空航天大学 A kind of unmanned aerial vehicle group frequency spectrum resource allocation method based on dynamic flight path
CN110336861A (en) * 2019-06-18 2019-10-15 西北工业大学 The unloading method for allocating tasks of mobile edge calculations system based on the double-deck unmanned plane

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017079623A1 (en) * 2015-11-06 2017-05-11 Massachusetts Institute Of Technology Dynamic task allocation in an autonomous multi-uav mission

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108632831A (en) * 2018-05-11 2018-10-09 南京航空航天大学 A kind of unmanned aerial vehicle group frequency spectrum resource allocation method based on dynamic flight path
CN110336861A (en) * 2019-06-18 2019-10-15 西北工业大学 The unloading method for allocating tasks of mobile edge calculations system based on the double-deck unmanned plane

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于stackelberg 多步博弈的无人机协同搜索路径规划;王瑞安等;《计算机工程与应用》;20190118;全文 *
面向异构无人机中继网络的负载均衡：一种分层博弈方法;杨婷婷等;《通信技术》;20181130;全文 *

Also Published As

Publication number Publication date
CN114205778A (en) 2022-03-18

Similar Documents

Publication Publication Date Title
CN111800828B (en) Mobile edge computing resource allocation method for ultra-dense network
CN109729528B (en) D2D resource allocation method based on multi-agent deep reinforcement learning
CN112512056B (en) Multi-objective optimization calculation unloading method in mobile edge calculation network
CN111182570B (en) User association and edge computing unloading method for improving utility of operator
CN108632831B (en) Unmanned aerial vehicle cluster frequency spectrum resource allocation method based on dynamic flight path
Pachler et al. Allocating power and bandwidth in multibeam satellite systems using particle swarm optimization
CN110049566B (en) Downlink power distribution method based on multi-unmanned-aerial-vehicle auxiliary communication network
CN113377533A (en) Dynamic computation unloading and server deployment method in unmanned aerial vehicle assisted mobile edge computation
CN108848535B (en) Sharing mode-oriented fog computing environment resource allocation method
Mahmud et al. Adaptive reinforcement learning framework for NOMA-UAV networks
CN115499921A (en) Three-dimensional trajectory design and resource scheduling optimization method for complex unmanned aerial vehicle network
CN115866787A (en) Network resource allocation method integrating terminal direct transmission communication and multi-access edge calculation
CN114142908B (en) Multi-unmanned aerial vehicle communication resource allocation method for coverage reconnaissance task
CN113490219B (en) Dynamic resource allocation method for ultra-dense networking
CN114205778B (en) Heterogeneous task-oriented unmanned aerial vehicle cluster cooperative target selection method
CN114051252A (en) Multi-user intelligent transmitting power control method in wireless access network
CN113038612A (en) Cognitive radio power control method based on deep learning
CN115811788B (en) D2D network distributed resource allocation method combining deep reinforcement learning and unsupervised learning
CN115038155B (en) Ultra-dense multi-access-point dynamic cooperative transmission method
CN115361453B (en) Load fair unloading and migration method for edge service network
Jiang et al. MARS: A DRL-based Multi-task Resource Scheduling Framework for UAV with IRS-assisted Mobile Edge Computing System
CN116074974A (en) Multi-unmanned aerial vehicle group channel access control method under layered architecture
CN115765826A (en) Unmanned aerial vehicle network topology reconstruction method for on-demand service
CN115529632A (en) Task unloading and resource allocation method facing edge computing
CN110149608B (en) DAI-based resource allocation method for optical wireless sensor network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant