CN116610144A - Unmanned plane collaborative dynamic task allocation method based on expansion consistency packet algorithm - Google Patents
Unmanned plane collaborative dynamic task allocation method based on expansion consistency packet algorithm Download PDFInfo
- Publication number
- CN116610144A CN116610144A CN202310579995.0A CN202310579995A CN116610144A CN 116610144 A CN116610144 A CN 116610144A CN 202310579995 A CN202310579995 A CN 202310579995A CN 116610144 A CN116610144 A CN 116610144A
- Authority
- CN
- China
- Prior art keywords
- task
- unmanned aerial
- aerial vehicle
- subtask
- tasks
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 48
- 238000004891 communication Methods 0.000 claims abstract description 14
- 238000010276 construction Methods 0.000 claims abstract description 6
- 239000011159 matrix material Substances 0.000 claims description 43
- 238000011156 evaluation Methods 0.000 claims description 16
- 230000008878 coupling Effects 0.000 claims description 12
- 238000010168 coupling process Methods 0.000 claims description 12
- 238000005859 coupling reaction Methods 0.000 claims description 12
- 239000000446 fuel Substances 0.000 claims description 10
- 238000003780 insertion Methods 0.000 claims description 4
- 230000037431 insertion Effects 0.000 claims description 4
- 238000012790 confirmation Methods 0.000 claims description 3
- 230000004083 survival effect Effects 0.000 claims description 3
- 230000002194 synthesizing effect Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 abstract description 8
- 230000000903 blocking effect Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 21
- 230000000875 corresponding effect Effects 0.000 description 13
- 230000009471 action Effects 0.000 description 8
- 230000008901 benefit Effects 0.000 description 6
- 230000007246 mechanism Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000012216 screening Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000035484 reaction time Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000035899 viability Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/10—Simultaneous control of position or course in three dimensions
- G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
- G05D1/104—Simultaneous control of position or course in three dimensions specially adapted for aircraft involving a plurality of aircrafts, e.g. formation flying
-
- 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
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an unmanned aerial vehicle collaborative dynamic task allocation method based on an extended consistency packet algorithm, which comprises the following steps: establishing a heterogeneous unmanned aerial vehicle dynamic task collaborative distribution model; constructing a task executable list according to performance conditions and task time sequence constraints of the heterogeneous unmanned aerial vehicle; each unmanned aerial vehicle performs bidding on tasks in the task executable list by using a consistency package algorithm to complete the construction of a local task sequence; the adjacent unmanned aerial vehicles carry out consistency negotiation on the conflicting task allocation schemes, consensus is achieved on task allocation results, and information in key information matrixes of unmanned aerial vehicles in the clusters is updated; when an emergency situation is faced, reassigning tasks based on a partial re-planning strategy; and circularly executing until the task allocation sequence of the unmanned aerial vehicle is not changed any more, and outputting a final task allocation scheme. The unmanned aerial vehicle collaborative dynamic task allocation method solves the problems that the existing method is easy to cause communication network blocking and large in calculation consumption.
Description
Technical Field
The invention belongs to the technical field of task allocation, and particularly relates to an unmanned aerial vehicle collaborative dynamic task allocation method based on an extended consistency packet algorithm.
Background
Compared with a manned plane, the unmanned plane (Unmanned Aerial Vehicles, UAV) has the characteristics of small volume, high flexibility, low cost, high viability, easy operation and the like, and plays an important role in tasks such as target reconnaissance, target striking, damage evaluation and the like. In a task environment with increasingly complex constraint conditions, the high robustness and the high task completion degree of the multi-unmanned aerial vehicle collaborative execution task have the advantages which are incomparable with those of the single unmanned aerial vehicle, and the defects that the single unmanned aerial vehicle is single in function and easy to be interfered by the outside are well overcome.
The multi-unmanned aerial vehicle multi-task cooperation aims to find an optimal task allocation method, and system performance is improved through multi-unmanned aerial vehicle cooperation under the restriction of task time constraint, position constraint, task type constraint and the like, which is essentially a multi-target optimization problem. The quality of unmanned aerial vehicle task allocation directly influences the execution efficiency and success rate of tasks, and indirectly influences the overall situation of the two parties. In addition, the actual task environment is dynamically changed continuously, and the occurrence of uncertain events such as damage to new tasks and unmanned aerial vehicle intelligent bodies can lead the original task allocation scheme to not meet task requirements. Therefore, research on the unmanned aerial vehicle cluster dynamic task allocation algorithm with higher efficiency and intelligence has important practical significance for development of unmanned aerial vehicles.
At present, students at home and abroad apply a large number of task allocation algorithms, and the algorithms can be summarized into three categories. One is a mathematical programming method, such as a mixed integer linear programming method, a hungarian algorithm, etc. However, when these algorithms are applied to solve large-scale problems, the solution space increases exponentially with the increase of the model, and the calculation amount increases, so that it becomes difficult to solve the problems while satisfying the real-time requirements. One is an intelligent optimization algorithm, such as a particle swarm optimization algorithm, a genetic algorithm, etc. The algorithm is generally used for solving the problem of centralized task allocation, has low complexity and good global characteristic, has high requirements on a central node, has poor system robustness, and is suitable for the collaborative execution of tasks of a short-distance and small-scale unmanned aerial vehicle system. In the process of executing tasks by the unmanned aerial vehicle cluster, high dynamic performance, complexity and uncertainty often exist, and the characteristic of striking damage exists, so that the algorithms are not suitable for dynamic task allocation with extremely high real-time requirements. The other is a market machine method, such as a contract net algorithm, an auction algorithm, etc. The contract net algorithm carries out bidding auction on the task, the unmanned aerial vehicle meeting the requirements sends a bid, and finally, proper unmanned aerial vehicle bid is selected to obtain the execution right of the task, so that the dynamic task allocation is realized. This approach is simple and feasible but convergence is difficult to ensure. The auction algorithm is that the unmanned aerial vehicle regards the task as an auction item, a certain task is auctioned according to a bidding strategy and an profit function, and dynamic task allocation is realized in a manner of continuous auction of multiple unmanned aerial vehicles. The process is a single-round single-bid-winning process, so when a plurality of new targets appear, a plurality of rounds of auctions are needed to determine a task allocation scheme, and the number of communication times among unmanned aerial vehicles is increased by the process of the plurality of rounds of auctions, so that the reaction time is increased.
The consistency package algorithm (CBBA) is a distributed algorithm developed on the basis of a contract net algorithm, and consists of two stages of unmanned aerial vehicle local task sequence construction and unmanned aerial vehicle conflict resolution. The algorithm can rapidly complete conflict-free allocation of tasks, but due to the constraint of task time windows, part of unmanned aerial vehicles have no opportunity to select tasks far away from themselves for execution, so that the danger of unmanned aerial vehicles is increased. The task allocation method mentioned above can achieve good effect in solving some task allocation problems, but in task allocation of unmanned aerial vehicle clusters, as the number of unmanned aerial vehicles and the target number increase, the complexity of calculation also increases sharply, so in task allocation of unmanned aerial vehicle cluster technology, conventional methods cannot complete task allocation in a short time.
In the prior art, a communication topological structure of complete connection is required, a large amount of data is transmitted, and the problem of communication blockage is easily caused; the problem of time sequence dynamic task allocation of the heterogeneous unmanned aerial vehicle cluster is difficult to solve; moreover, the method is extremely easy to fall into the dead loop of the re-planning state, and has poor instantaneity.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides an unmanned aerial vehicle collaborative dynamic task allocation method based on an extended consistency packet algorithm, solves the problems that the existing method is easy to cause communication network blocking and large in calculation consumption, introduces task time window constraint and partial re-planning mechanism, and solves the problem that the existing method is poor in solving instantaneity. The technical problems to be solved by the invention are realized by the following technical scheme:
the invention provides an unmanned aerial vehicle collaborative dynamic task allocation method based on an extended consistency packet algorithm, which comprises the following steps:
s1: according to task environment and task demand information, establishing a heterogeneous unmanned aerial vehicle dynamic task collaborative distribution model formed by a performance index function and constraint conditions;
s2: constructing a task executable list according to performance conditions and task time sequence constraints of the heterogeneous unmanned aerial vehicle;
s3: each unmanned aerial vehicle utilizes a consistency package algorithm to bid on tasks in the task executable list, and the construction of a local task sequence of each unmanned aerial vehicle is completed;
s4: establishing communication between adjacent unmanned aerial vehicles to interact bidding information of the two parties, carrying out consistency negotiation on a conflicting task allocation scheme, achieving consensus on a task allocation result and updating information in a key information matrix of unmanned aerial vehicles in a cluster;
s5: when an emergency situation is faced, reassigning tasks based on a partial re-planning strategy;
s6: and (3) circularly executing S3 to S5 until the task allocation sequence of the unmanned aerial vehicle is not changed any more, and outputting a final task allocation scheme.
In one embodiment of the present invention, the S1 includes:
s1.1: acquiring an unmanned aerial vehicle set and a target task set according to the task environment, the unmanned aerial vehicle types and number and the task types and number;
s1.2: comprehensively considering task execution income and fuel consumption factors, and designing a performance index function;
s1.3: setting multi-machine coordination constraint, unmanned aerial vehicle voyage constraint and task load balancing constraint according to the type and the number of loads carried by the unmanned aerial vehicle and the performance of the unmanned aerial vehicle, and setting task time sequence constraint and task time window constraint according to the execution time sequence among tasks and the task time;
s1.4: and synthesizing the performance index function and each constraint condition, and constructing a heterogeneous unmanned aerial vehicle dynamic task collaborative distribution model.
In one embodiment of the present invention, the S1.2 includes:
s1.21: obtaining the success probability of the ith unmanned aerial vehicle to complete the jth task:
wherein, target confirmation probability representing unmanned aerial vehicle in completing scout task stage,/>Representing the damage probability of the unmanned plane to the target in the stage of performing the striking task>The survival probability of the unmanned aerial vehicle is represented;
s1.22: comprehensively considering the overall income of executing tasks and the flight fuel consumption of all unmanned aerial vehicles, and designing a scoring function as follows:
wherein DeltaL ij Is unmanned plane V i Mission allocation plan flight path value, f i Is unmanned plane V i The fuel consumption generated per kilometer range, lambda is a value attenuation factor, p i Is unmanned plane V i Value of the path sequence set j0 Representing Task j Value of (1) j Representing Task j T is the current value of ij Represents the moment when the unmanned aerial vehicle i performs task j, u (t ij ) Is a binary variable representing t ij Whether a time window of a task is satisfied;
s1.23: the obtained performance index function is:
wherein, representing unmanned plane V as decision variable in multi-plane cooperative constraint i Whether or not to be allocated to execute Task j ，N V Represents the total number of heterogeneous unmanned aerial vehicles, N T Representing the total number of targets, N C Representing the total number of tasks.
In one embodiment of the invention, the task time window constraint is:
wherein, respectively represent the target T j A time window of a scout task, a hit task and a damage evaluation task, wherein a subscript start represents a start time of the time window, and a subscript end represents an end time;
the task timing constraints are:
in one embodiment of the present invention, the S2 includes:
s2.1: constructing a subtask relation matrix and a corresponding time sequence interval matrix according to the coupling constraint relation among tasks and the types and the quantity of loads required;
s2.2: and constructing a task executable list of the unmanned aerial vehicle according to the load information of the unmanned aerial vehicle, the subtask relation matrix and the corresponding time sequence interval matrix.
In one embodiment of the present invention, the S2.1 includes:
defining the subtask relationship matrix as a square matrix d={d wq W=1, 2,3; q=1, 2,3, where d wq Representing a constraint of coupling relation between subtask w and subtask q, d wq =1 indicates that there is a timing constraint between subtask w and subtask q, d wq =0 indicates that there is no timing constraint relationship between subtask w and subtask q;
obtaining a minimum time interval Deltat between subtask w and subtask q min ：
Wherein T is a subtask time sequence interval matrix, T wq Representing a time interval relationship between subtasks w and q;
for N T The target respectively and sequentially carries out a reconnaissance task, a striking task and a damage evaluation task, and the corresponding subtask relation matrix is as follows:
the corresponding subtask time sequence interval matrix is as follows:
wherein Δt is min The minimum time required to complete the previous subtask before the next subtask is executed.
In one embodiment of the present invention, the S3 includes:
s3.1: initializing a key information matrix for storing task allocation information, wherein the key information matrix comprises a task package set, a path sequence set, a winning party bid set, a timestamp set, a task execution time sequence and a task time window list, the task execution time sequence is used for storing execution time of the unmanned aerial vehicle on a corresponding task, and the task time window list is used for storing a task executable time window;
s3.2: according to a sequence greedy principle, each unmanned aerial vehicle independently performs bidding on tasks in a task executable list of the unmanned aerial vehicle, and adds the tasks into a task package of the unmanned aerial vehicle;
s3.3: step S3.2 is looped until the current drone reaches the maximum number of executable tasks.
In one embodiment of the present invention, the S3.2 includes:
each unmanned plane performs bidding on tasks in an executable list of the task by taking the increment of the maximum self task gain as a selection principle, and a current path set p i Respectively inserting the task j into all positions in the system, calculating the gain of the score function after insertion, and comparing, if the task j is inserted, the unmanned plane V i And adding the task j into the task package of the unmanned aerial vehicle i when the gain of the scoring function is large, and updating the key information matrix.
In one embodiment of the present invention, the S4 includes:
s4.1: the adjacent unmanned aerial vehicle carries out consistency negotiation according to a consensus principle by utilizing the correlation values in the winning party set, the winning party bid set and the timestamp set so as to resolve conflict;
s4.2: and after the conflict resolution is completed, updating the information in the key information matrix of the unmanned aerial vehicle in the cluster.
In one embodiment of the present invention, the S5 includes:
s5.1: the method comprises the steps that task information is clarified, a ground station obtains residual task information or new task information of the crash unmanned aerial vehicle, the residual task information or the new task information comprises the position, the type and the load requirement of a new target task, and broadcasting is carried out to an unmanned aerial vehicle cluster;
s5.2: and according to the new task information and the self state, each unmanned aerial vehicle distributes new tasks based on the partial re-planning distribution strategy, and adjusts an original task distribution scheme.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention corrects the dynamic task allocation model, comprehensively considers constraint conditions such as load types and quantity of unmanned aerial vehicles, multi-machine cooperation and load balancing, task time sequence, time window and the like, builds an allocation model capable of effectively solving the problems of the large-scale unmanned aerial vehicle and the target tasks by taking the value gains of the tasks and the flying fuel consumption as evaluation indexes, ensures that all the target tasks can be effectively executed, and is a more practical advantage of the model.
2. According to the invention, the task executable list is constructed, the unmanned aerial vehicle adds the tasks meeting the self-load resource requirements and conforming to the coupling constraint into the task executable list according to the self-load information, so that the heterogeneous and task coupling constraint problems of the unmanned aerial vehicle are solved, one round of task screening is performed before task bidding, and unnecessary computing resource consumption caused by the unmanned aerial vehicle on all task bidding is reduced.
3. The invention introduces a partial re-planning allocation strategy, and re-allocates the left-over task and the sudden new target task of the crashed unmanned aerial vehicle when the emergency situation is faced, such as the crash of the unmanned aerial vehicle or the occurrence of the new target task. The dynamic adjustment of the task allocation scheme is carried out on the basis of the original task allocation scheme, so that the communication consumption and the calculation resource consumption caused by the complete re-planning allocation strategy are reduced, and the instantaneity of the task re-allocation system is improved.
The present invention will be described in further detail with reference to the accompanying drawings and examples.
Drawings
Fig. 1 is a flowchart of an unmanned aerial vehicle collaborative dynamic task allocation method based on an extended consistency packet algorithm provided by an embodiment of the invention;
fig. 2 is a flowchart of an extended consistency packet algorithm based on a partial re-planning mechanism according to an embodiment of the present invention.
Detailed Description
In order to further explain the technical means and effects adopted by the invention to achieve the preset aim, the unmanned aerial vehicle collaborative dynamic task allocation method based on the extended consistency packet algorithm provided by the invention is described in detail below with reference to the accompanying drawings and the specific embodiments.
The foregoing and other features, aspects, and advantages of the present invention will become more apparent from the following detailed description of the preferred embodiments when taken in conjunction with the accompanying drawings. The technical means and effects adopted by the present invention to achieve the intended purpose can be more deeply and specifically understood through the description of the specific embodiments, however, the attached drawings are provided for reference and description only, and are not intended to limit the technical scheme of the present invention.
It should be noted that in this document relational terms such as first and second, and the like are 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. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that an article or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in an article or apparatus that comprises the element.
Referring to fig. 1 and fig. 2, fig. 1 is a flowchart of an unmanned plane collaborative dynamic task allocation method based on an extended consistency packet algorithm according to an embodiment of the present invention, and fig. 2 is a flowchart of an extended consistency packet algorithm based on a partial re-planning mechanism according to an embodiment of the present invention. The task allocation method comprises the following steps:
s1: and establishing a heterogeneous unmanned aerial vehicle dynamic task collaborative distribution model formed by a performance index function and constraint conditions according to the task environment and task demand information.
Specifically, S1 of the present embodiment includes:
s1.1: and obtaining an unmanned aerial vehicle set and a target task set according to the task environment, the unmanned aerial vehicle types and number and the task types and number.
In the present embodiment, there are N in the space V Cooperative task execution of unmanned aerial vehicle frames toRepresentation, comprising two types of drones: reconnaissanceUnmanned aerial vehicle with flight speed v c The method has the advantages that the method does not carry attack load, can execute reconnaissance tasks and damage evaluation tasks, and has high task success probability; fighter unmanned plane with flight speed v f The attack load is carried, and besides the scout task and the damage evaluation task can be executed, the striking task can be additionally executed, but the success probability of the scout task and the damage evaluation task is lower. With N T Target, in->Indicating that the target moving speed is v t . For N T The target respectively and sequentially carries out reconnaissance task, striking task and damage evaluation task, and the total number of the tasks to be executed is N C ＝3N T ToIndicating that the fixed value of each task is +.>
S1.2: and comprehensively considering factors such as task execution income, fuel consumption and the like, designing a performance index function, wherein a task execution time window and a time attenuation factor are introduced, ensuring that the performance index function is a marginal gain decreasing function, and being capable of stably converging. Meanwhile, increasing the initial value of the task ensures that the performance index function is always positive, and all target tasks can be effectively distributed and executed all the time.
Specifically, assume thatThe success probability of completing the j-th task for the i-th unmanned aerial vehicle can be expressed as:
wherein, target confirmation probability representing unmanned aerial vehicle in completing scout task stage,/>Representing the damage probability of the unmanned plane to the target in the stage of performing the striking task>The survival probability of the unmanned aerial vehicle is represented.
Comprehensively considering the overall income of executing tasks and the flight fuel consumption of all unmanned aerial vehicles, and designing a scoring function as follows:
wherein DeltaL ij Is unmanned plane V i Mission allocation plan flight path value, f i Is unmanned plane V i The fuel consumption generated per kilometer range, lambda is a value attenuation factor, p i Is unmanned plane V i Value of the path sequence set j0 Representing Task j Value of (1) j Representing Task j T is the current value of ij Represents the moment when the unmanned aerial vehicle i performs task j, u (t ij ) Is a binary variable representing t ij If the time window of the task is satisfied, the task cannot be executed if the time exceeds the time window of the task, u (t ij ) The expression of (2) is:
wherein t is start Time window start time, t, representing task j end Indicating the time window end time of task j.
In order to make unmanned plane V i As many bidding tasks as possible, need to ensure S i (p i )>0, thus the initial Value of the task j0 The following should be satisfied:
thus, the performance index function is:
wherein, representing decision variables in a multi-machine collaboration constraint, the expression of which will be described in detail below.
S1.3: according to the type and the number of the load carried by the unmanned aerial vehicle and the performance of the unmanned aerial vehicle, multi-machine coordination constraint, unmanned aerial vehicle range constraint and task load balancing constraint are set, and task time window constraint and task time sequence constraint are set according to the execution time sequence among tasks and task time.
In this embodiment, the constraint condition setting process is as follows:
(1) Task time window constraints and task timing constraints:
it should be noted that, the execution of three tasks (i.e., the scout task, the damage evaluation task, and the hit task) of each target has strict timing and time requirements, and the scout task must be completed first, then the hit task, and finally the evaluation task. Setting target T j Is a time window for each task:
wherein, respectively represent the target T j A time window of a scout task, a hit task and a damage assessment task, a subscript start represents a start time of the time window, and a subscript end represents an end time. The task timing constraint is:
(2) Multi-machine cooperative constraint:
in order to avoid the problem of coordination when unmanned aerial vehicle clusters execute tasks, such as repeated execution or missing of targets, the number of hit tasks exceeding the number of fighter ammunition, and the like, multi-machine coordination constraint needs to be added when a heterogeneous unmanned aerial vehicle dynamic task coordination distribution model is constructed, and the key points are as follows:
(1) each task only needs one unmanned aerial vehicle to execute, so each task can be completed once only, and the unmanned aerial vehicle is provided withAs a decision variable, its value satisfies:
the constraint may be expressed as follows:
(2) aiming at the striking task, the number of tasks distributed by the unmanned aerial vehicle is required to be less than or equal to the maximum ammunition number, and m is set i Is unmanned plane V i (V i E V), the constraint can be expressed as:
(3) all tasks need to be performed and constraints can be expressed as:
(3) Unmanned aerial vehicle voyage constraint:
arbitrary unmanned plane V i Is smaller than its maximum flight rangeThe unmanned voyage constraint is therefore expressed as:
wherein L is (i,j) Representing unmanned plane V i Fly from last Task point to current Task j Number of voyages at the location.
(4) Task load balancing constraints:
in task allocation, the situation that the task allocation is too concentrated on a certain unmanned aerial vehicle sometimes occurs, so that the resource utilization rate of the unmanned aerial vehicle is low, and the overall efficiency of completing tasks by the unmanned aerial vehicle cluster is low. In view of this, the embodiments of the present invention introduce task load balancing constraints. Definition unmanned plane V i Is load as task load i The average task load of the unmanned aerial vehicle cluster isThe computational expression is:
unmanned aerial vehicle V i The task load rate of (1) is:
if the task load rate is smaller than e, the user-defined constant e indicates that the unmanned aerial vehicle has fewer tasks distributed at present, and other tasks can be continuously executed; otherwise, the drone may not continue to perform other tasks.
S1.4: and constructing a heterogeneous unmanned aerial vehicle dynamic task collaborative distribution model by integrating the performance index function and each constraint condition, namely a combined optimization model of a multi-frame heterogeneous unmanned aerial vehicle collaborative dynamic time sequence task distribution problem.
S2: and constructing a task executable list (can do list) according to the performance condition of the heterogeneous unmanned aerial vehicle and task time sequence constraint.
In the present embodiment, step S2 includes the steps of:
s2.1: and constructing a subtask relation matrix and a corresponding time sequence interval matrix according to the coupling constraint relation among the tasks and the types and the quantity of the required loads.
Specifically, the target j may be divided into three subtasks, which respectively require different types of load resources, and each subtask only requires one load resource. The subtask allocation has a coupling constraint relation, and a subtask relation matrix is defined as a square matrix D= { D for describing the coupling constraint relation wq W=1, 2,3; q=1, 2,3, where d wq Representing a coupling relationship constraint between subtask w and subtask q, d wq =1 indicates that there is a timing constraint between subtask w and subtask q, d wq =0 indicates that there is no timing constraint relationship between subtask w and subtask q. When there is a timing relationship constraint between subtask w and subtask q, there is a minimum time interval Δt between the start times of w and q min The method comprises the following steps:
wherein T is a subtask time sequence interval matrix, T wq Representing the time interval relationship between subtask w and subtask q.
For N T Each target sequentially carries out a reconnaissance task, a striking task and a damage evaluation task, and the subtask relation matrix corresponding to each target is as follows:
the corresponding subtask time sequence interval matrix is as follows:
wherein Δt is min The minimum time required to complete the previous subtask before the next subtask is executed.
S2.2: and constructing a task executable list of the unmanned aerial vehicle according to the load information of the unmanned aerial vehicle, the subtask relation matrix and the corresponding time sequence interval matrix.
Specifically, unmanned plane V i Firstly, determining whether Task can be executed or not according to self-load information j Adding tasks meeting self-load resource requirements to a task executable list, e.g. assuming unmanned aerial vehicle V i Is a reconnaissance unmanned plane, and can only execute reconnaissance and damage evaluation tasks based on load capacity constraint, so that unmanned plane V i Can only add the two types of tasks to the task executable list of the unmanned plane V i Is a capability constraint of (a). Secondly, only unmanned plane V i Before the previous task of the same target has been added to the task executable list of (a) to add the next task of the target, e.g. unmanned aerial vehicle V in the case of a scout unmanned aerial vehicle i Only after the corresponding reconnaissance task is distributed on the same target, the damage evaluation task can be added into the task executable list, so that the time sequence constraint of the task is met. Finally, unmanned plane V i In the subsequent self task package construction stage, only the tasks in the task executable list are subjected to bidding, so that the unmanned aerial vehicle V is reduced i Unnecessary computations for bidding on all tasks.
S3: and each unmanned aerial vehicle asynchronously utilizes a consistency packet algorithm to bid on the tasks in the task executable list, and the local task sequence construction of each unmanned aerial vehicle is completed.
Specifically, S3 of the present embodiment includes the following steps:
s3.1: initializing a key information matrix for storing task allocation information, wherein the key information matrix comprises a task package set, a path sequence set, a winning party bid set, a time stamp set and a task execution time sequenceτ i And task time window list u i 。
Illustratively, in unmanned plane V i The following are examples:
(1) The task package set b is represented as:
wherein b in ∈T(n＝1,2,...,|b i I) represents unmanned plane V i Task number obtained by current auction, |b i The number of elements of the set, i.e., the number of tasks, is denoted by b 13 For example, =2, represents the unmanned plane V 1 Task 3 obtained by auction is Task 2 The method comprises the steps of carrying out a first treatment on the surface of the The front element in the set is assigned to the unmanned aerial vehicle earlier than the rear element, for example: task b i1 Specific task b i2 Is first allocated to unmanned plane V i The method comprises the steps of carrying out a first treatment on the surface of the In particular, the method comprises the steps of,representing unmanned plane V i No auction to the task.
(2) The path order set p is expressed as:
wherein p is in ∈T(n＝1,2,...,|p i I) represents unmanned plane V i The order in which tasks are currently ready to be performed is p 13 For example, =2, represents the unmanned plane V 1 Task 3 ready to execute is Task 2 The method comprises the steps of carrying out a first treatment on the surface of the The former element of the collection is performed by the drone earlier than the latter element, e.g. task p i1 Than task p i2 Unmanned aerial vehicle V i Executing; therefore, the element sequence in the path sequence set p represents the execution sequence of the unmanned aerial vehicle to the tasks in the task package.
(3) The winning set z is expressed as:
wherein z is in ∈V(n＝1,2,...,|z i I) represents unmanned plane V i Consider that a certain unmanned aerial vehicle can execute Task n In z 13 For example, =2, represents the unmanned plane V 1 Consider unmanned plane V 2 Task may be performed 3 The method comprises the steps of carrying out a first treatment on the surface of the In particular, the method comprises the steps of,representing unmanned plane V i Task is considered n Not being bid upon by any drone.
(4) The winner bid set y is expressed as:
wherein y is in (n＝1,2,...,|y i I) represents a bid to get Task n Is the highest bid of (2); specifically, y in =0 denotes Task n And is not auctioned.
(5) The set of timestamps s is denoted as:
wherein s is in (n＝1,2,...,N V ) Representing unmanned plane V i Accepted from unmanned plane V n The time of the latest information, the greater the value thereof, represents the unmanned plane V i The more recent the information received.
In a dynamic environment, the targets are mostly time-sensitive targets, and the method has high maneuverability. Thus, the division algorithm itself uses b i 、p i 、z i And y i Waiting list for storing unmanned aerial vehicle V i In addition to bidding information, a task execution time sequence tau is adopted i To store the execution time of the unmanned aerial vehicle on the corresponding task and a task time window list u i To store a task executable time window. Wherein, τ in (n＝1,2,...,|τ i i) is unmanned plane V i Executing task p according to its own path sequence set in The starting time is therefore τ i1 <τ i2 <...<τ in ；/>|u ij |＝end ij -start ij ，u ij Representing unmanned plane V i Task execution Task j Is performed for a window of time, unmanned plane V i Task execution only during this time frame j The calculation is valid.
S3.2: according to the order greedy principle, each unmanned aerial vehicle independently performs bidding on tasks in the task executable list, and adds the tasks into the task package.
Illustratively, each unmanned aerial vehicle performs bidding on the effective tasks meeting the time sequence constraint and the load resource constraint, namely the tasks in the self task executable list, by taking the increment of the maximum self task income as a selection principle. For the current path set p i Respectively inserting the task j and calculating the gain of the score function after insertion and comparing, wherein the formula is constructed as follows:
wherein c ij (p i ) Gains of the scoring function of the unmanned plane i after adding the task j into the task executable list, namely gains obtained by executing the task;representing insertion of task j into p i An nth position of (a); />Representing unmanned plane V i Sequential set p along path i The benefits of executing the task, i.e. the total benefits obtained by task allocation, are given by the formula at step 1.
Unmanned plane V after task j is inserted i The task j is added to the unmanned plane V by the gain of the scoring function i And updates the key information matrix.
Unmanned plane V after task j is inserted i The task j is added to the unmanned plane V by the gain of the scoring function i Task package set b i Is the last position in the series. b i Is constructed as follows:
inserting task j into p i The position n at which the unmanned aerial vehicle performs the task to obtain the maximum score. P is p i Is constructed as follows:
then update the winner set z ij And winning bid set y ij Let y ij ＝c ij ，z ij =i; updating tau i Unmanned aerial vehicle V is stored i Execution time on task j.
S3.3: and (3) circulating the step S3.2 until the unmanned aerial vehicle V i Up to the maximum number N of executable tasks i 。
S4: communication is established between adjacent unmanned aerial vehicles to interact bidding information of the two parties, consistency negotiation is carried out on conflicting task allocation schemes, and consensus is achieved on task allocation results. The bidding information includes a winning set, a winning bid set, and a time stamp list for each drone.
In this embodiment, the present step specifically includes the following steps:
s4.1: using winning drone matrix z i Winning bid-competing matrix y i And a timestamp list s i Information according to consensusConsistency negotiations are conducted in principle.
Specifically, during the conflict resolution phase, the drone utilizes a list of timestamps s i To determine the latest information received. Unmanned plane V k Represents bidding information sender, unmanned plane V i Representing bidding information receiver, every time unmanned plane V i Receiving from adjacent unmanned aerial vehicle V k When the information is marked with the following time stamp:
wherein g ik =1 represents the unmanned plane V i With adjacent unmanned plane V k A communication link exists between them, otherwise g ik =0, in particular g ii ＝1；T r The time when the bidding information is received for the unmanned aerial vehicle i. If unmanned plane V i With unmanned aerial vehicle V k No communication connection exists between the two unmanned aerial vehicles, multi-hop transfer is needed through other unmanned aerial vehicles,indicating the time of receipt of the information after delivery.
The receiver determines what action to take by itself through comprehensive judgment according to the relevant values in the winning party set, the winning party bid set and the timestamp set sent by the sender. Unmanned plane V i Task to Task j The following three actions can be taken:
the expression represents unmanned plane V i Updating information in winning party set z and winning party bid set y corresponding to different actions, wherein y ij Representing unmanned plane V i Stored Task-for-Task j Highest bid of y kj Representation and unmanned plane V i Unmanned plane V for communication k Stored Task-for-Task j Is the highest bid of (2); z ij Representing unmanned plane V i Consider bidding Task j Winning party, z kj Representing unmanned plane V k Consider bidding Task j Winner of (a).
Unmanned plane V i Upon receiving the unmanned aerial vehicle V k Task related Task j After the information of (2), corresponding actions are taken based on the conflict negotiation resolution consensus rule shown in table 1. The values in the first two columns in Table 1 are respectively unmanned plane V k And V is equal to i Evaluation of identified bidding Task j The third column indicates the winner of drone V i Actions that should be performed, wherein the default action is leave.
Table 1 conflict negotiation resolution consensus rule table
S4.2: and after the conflict resolution is completed, updating the information in the key information matrix of the unmanned aerial vehicle in the cluster.
Specifically, if unmanned plane V i Winning set z i And winning bid set y i After negotiation based on the consensus mechanism of table 1, updating occurs, and the winning party set z is selected from the task package set and the path sequence set of all unmanned aerial vehicles i And winning bid set y i The affected task after the value update and the subsequent tasks are released and reset.
S5: in the face of an emergency, such as a crash of an unmanned aerial vehicle or the appearance of a new target task, the task is reassigned based on a partial re-planning strategy.
In this embodiment, step S5 includes:
s5.1: and (3) defining task information, acquiring residual task information or new target task information of the crash unmanned aerial vehicle by the ground station, including positions, types and load requirements, and broadcasting to the unmanned aerial vehicle clusters.
S5.2: and according to the new task information and the self state, each unmanned aerial vehicle distributes new tasks based on the partial re-planning distribution strategy, and adjusts an original task distribution scheme.
Illustratively, each unmanned aerial vehicle first checks whether the task package of itself is full, if the task package is not full, repeat steps S3 to S4, add new task on the basis of original task allocation scheme; if the self task package is full, the unmanned aerial vehicle willingly bidding releases the lowest profit value from the current task package when the task package is constructedAnd (5) keeping the other tasks unchanged. />The magnitude of the value can be determined by comprehensively considering the new task number and the response speed (namely, the algorithm running speed) requirement. When new tasks frequently occur and the unmanned aerial vehicle cluster system needs to realize conflict-free task allocation before the next new task occurs, the +.>The numerical value is smaller; when the new task value is very high, the cooperation capability of the unmanned aerial vehicle can be improved by releasing more tasks. And then, repeating S3 to S4, reallocating the new target task, and adjusting the original allocation scheme. />
S6: and (3) circularly executing S3 to S5 until the task allocation sequence of the unmanned aerial vehicle is not changed any more, and outputting a final task allocation scheme which comprises an unmanned aerial vehicle task execution path diagram, a task allocation schedule and the like.
The unmanned aerial vehicle collaborative dynamic task allocation method based on the expansion consistency packet algorithm solves the problems that the existing method is easy to cause communication network blockage and large in calculation consumption, introduces task time window constraint and partial re-planning mechanism, and solves the problem that heterogeneous unmanned aerial vehicle clusters solve time sequence dynamic tasks in poor real-time. According to the invention, the task executable list is constructed, the unmanned aerial vehicle adds the tasks meeting the self-load resource requirements and conforming to the coupling constraint into the task executable list according to the self-load information, so that the heterogeneous and task coupling constraint problems of the unmanned aerial vehicle are solved, one round of task screening is performed before task bidding, and unnecessary computing resource consumption caused by the unmanned aerial vehicle on all task bidding is reduced.
In addition, the invention introduces a partial re-planning allocation strategy, and re-allocates the left-over task and the sudden new target task of the crashed unmanned aerial vehicle when an emergency situation exists, such as the crash of the unmanned aerial vehicle or the occurrence of the new target task. The dynamic adjustment of the task allocation scheme is carried out on the basis of the original task allocation scheme, so that the communication consumption and the calculation resource consumption caused by the complete re-planning allocation strategy are reduced, and the instantaneity of the task re-allocation system is improved.
A further embodiment of the present invention provides a storage medium, where a computer program is stored, where the computer program is configured to perform the steps of the unmanned aerial vehicle collaborative dynamic task allocation method based on the extended consistency packet algorithm described in the above embodiment. In a further aspect, the present invention provides an electronic device, including a memory and a processor, where the memory stores a computer program, and the processor, when calling the computer program in the memory, implements the steps of the unmanned plane collaborative dynamic task allocation method based on the extended consistency packet algorithm according to the above embodiment. In particular, the integrated modules described above, implemented in the form of software functional modules, may be stored in a computer readable storage medium. The software functional module is stored in a storage medium and includes instructions for causing an electronic device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) to perform part of the steps of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is a further detailed description of the invention in connection with the preferred embodiments, and it is not intended that the invention be limited to the specific embodiments described. It will be apparent to those skilled in the art that several simple deductions or substitutions may be made without departing from the spirit of the invention, and these should be considered to be within the scope of the invention.
Claims (10)
1. The unmanned aerial vehicle collaborative dynamic task allocation method based on the extended consistency packet algorithm is characterized by comprising the following steps of:
s1: according to task environment and task demand information, establishing a heterogeneous unmanned aerial vehicle dynamic task collaborative distribution model formed by a performance index function and constraint conditions;
s2: constructing a task executable list according to performance conditions and task time sequence constraints of the heterogeneous unmanned aerial vehicle;
s3: each unmanned aerial vehicle utilizes a consistency package algorithm to bid on tasks in the task executable list, and the construction of a local task sequence of each unmanned aerial vehicle is completed;
s4: establishing communication between adjacent unmanned aerial vehicles to interact bidding information of the two parties, carrying out consistency negotiation on a conflicting task allocation scheme, achieving consensus on a task allocation result and updating information in a key information matrix of unmanned aerial vehicles in a cluster;
s5: when an emergency situation is faced, reassigning tasks based on a partial re-planning strategy;
s6: and (3) circularly executing S3 to S5 until the task allocation sequence of the unmanned aerial vehicle is not changed any more, and outputting a final task allocation scheme.
2. The unmanned aerial vehicle collaborative dynamic task allocation method based on the extended consistency packet algorithm according to claim 1, wherein S1 comprises:
s1.1: acquiring an unmanned aerial vehicle set and a target task set according to the task environment, the unmanned aerial vehicle types and number and the task types and number;
s1.2: comprehensively considering task execution income and fuel consumption factors, and designing a performance index function;
s1.3: setting multi-machine coordination constraint, unmanned aerial vehicle voyage constraint and task load balancing constraint according to the type and the number of loads carried by the unmanned aerial vehicle and the performance of the unmanned aerial vehicle, and setting task time sequence constraint and task time window constraint according to the execution time sequence among tasks and the task time;
s1.4: and synthesizing the performance index function and each constraint condition, and constructing a heterogeneous unmanned aerial vehicle dynamic task collaborative distribution model.
3. The unmanned aerial vehicle collaborative dynamic task allocation method based on the extended consistency packet algorithm according to claim 2, wherein S1.2 comprises:
s1.21: obtaining the success probability of the ith unmanned aerial vehicle to complete the jth task:
wherein, target confirmation probability representing unmanned aerial vehicle in completing scout task stage,/>Representing the damage probability of the unmanned plane to the target in the stage of performing the striking task>The survival probability of the unmanned aerial vehicle is represented;
s1.22: comprehensively considering the overall income of executing tasks and the flight fuel consumption of all unmanned aerial vehicles, and designing a scoring function as follows:
wherein DeltaL ij Is unmanned plane V i Mission allocation plan flight path value, f i Is unmanned plane V i The fuel consumption generated per kilometer range, lambda is a value attenuation factor, p i Is unmanned plane V i Path sequence set, value j0 Representing Task j Value of (1) j Representing Task j T is the current value of ij Represents the moment when the unmanned aerial vehicle i performs task j, u (t ij ) Is a binary variable representing t ij Whether a time window of a task is satisfied;
s1.23: the obtained performance index function is:
wherein, representing unmanned plane V as decision variable in multi-plane cooperative constraint i Whether or not to be allocated to execute Task j ，N V Represents the total number of heterogeneous unmanned aerial vehicles, N T Representing the total number of targets, N C Representing the total number of tasks.
4. The unmanned aerial vehicle collaborative dynamic task allocation method based on the extended consistency packet algorithm according to claim 2, wherein the task time window constraint is:
wherein, respectively represent the target T j A time window of a scout task, a hit task and a damage evaluation task, wherein a subscript start represents a start time of the time window, and a subscript end represents an end time;
the task timing constraints are:
5. the unmanned aerial vehicle collaborative dynamic task allocation method based on the extended consistency packet algorithm according to claim 2, wherein S2 comprises:
s2.1: constructing a subtask relation matrix and a corresponding time sequence interval matrix according to the coupling constraint relation among tasks and the types and the quantity of loads required;
s2.2: and constructing a task executable list of the unmanned aerial vehicle according to the load information of the unmanned aerial vehicle, the subtask relation matrix and the corresponding time sequence interval matrix.
6. The unmanned aerial vehicle collaborative dynamic task allocation method based on the extended consistency packet algorithm according to claim 5, wherein S2.1 comprises:
defining the subtask relation matrix as a square matrix D= { D wq W=1, 2,3; q=1, 2,3, where d wq Representing a constraint of coupling relation between subtask w and subtask q, d wq =1 indicates that there is a timing constraint between subtask w and subtask q, d wq =0 indicates that there is no timing constraint relationship between subtask w and subtask q;
obtaining a minimum time interval Deltat between subtask w and subtask q min ：
Wherein T is a subtask time sequence interval matrix, T wq Representing a time interval relationship between subtasks w and q;
for N T The target respectively and sequentially carries out a reconnaissance task, a striking task and a damage evaluation task, and the corresponding subtask relation matrix is as follows:
the corresponding subtask time sequence interval matrix is as follows:
wherein Δt is min The minimum time required to complete the previous subtask before the next subtask is executed.
7. The unmanned aerial vehicle collaborative dynamic task allocation method based on the extended consistency packet algorithm according to claim 5, wherein S3 comprises:
s3.1: initializing a key information matrix for storing task allocation information, wherein the key information matrix comprises a task package set, a path sequence set, a winning party bid set, a timestamp set, a task execution time sequence and a task time window list, the task execution time sequence is used for storing execution time of the unmanned aerial vehicle on a corresponding task, and the task time window list is used for storing a task executable time window;
s3.2: according to a sequence greedy principle, each unmanned aerial vehicle independently performs bidding on tasks in a task executable list of the unmanned aerial vehicle, and adds the tasks into a task package of the unmanned aerial vehicle;
s3.3: step S3.2 is looped until the current drone reaches the maximum number of executable tasks.
8. The unmanned aerial vehicle collaborative dynamic task allocation method based on the extended consistency packet algorithm according to claim 7, wherein S3.2 comprises:
each unmanned plane performs bidding on tasks in an executable list of the task by taking the increment of the maximum self task gain as a selection principle, and a current path set p i Respectively inserting the task j into all positions in the system, calculating the gain of the score function after insertion, and comparing, if the task j is inserted, the unmanned plane V i And adding the task j into the task package of the unmanned aerial vehicle i when the gain of the scoring function is large, and updating the key information matrix.
9. The unmanned aerial vehicle collaborative dynamic task allocation method based on the extended consistency packet algorithm according to claim 7, wherein S4 comprises:
s4.1: the adjacent unmanned aerial vehicle carries out consistency negotiation according to a consensus principle by utilizing the correlation values in the winning party set, the winning party bid set and the timestamp set so as to resolve conflict;
s4.2: and after the conflict resolution is completed, updating the information in the key information matrix of the unmanned aerial vehicle in the cluster.
10. The unmanned aerial vehicle collaborative dynamic task allocation method based on the extended consistency packet algorithm according to claim 9, wherein S5 comprises:
s5.1: the method comprises the steps that task information is clarified, a ground station obtains residual task information or new task information of the crash unmanned aerial vehicle, the residual task information or the new task information comprises the position, the type and the load requirement of a new target task, and broadcasting is carried out to an unmanned aerial vehicle cluster;
s5.2: and according to the new task information and the self state, each unmanned aerial vehicle distributes new tasks based on the partial re-planning distribution strategy, and adjusts an original task distribution scheme.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310579995.0A CN116610144A (en) | 2023-05-22 | 2023-05-22 | Unmanned plane collaborative dynamic task allocation method based on expansion consistency packet algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310579995.0A CN116610144A (en) | 2023-05-22 | 2023-05-22 | Unmanned plane collaborative dynamic task allocation method based on expansion consistency packet algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116610144A true CN116610144A (en) | 2023-08-18 |
Family
ID=87677697
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310579995.0A Pending CN116610144A (en) | 2023-05-22 | 2023-05-22 | Unmanned plane collaborative dynamic task allocation method based on expansion consistency packet algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116610144A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117873137A (en) * | 2024-03-12 | 2024-04-12 | 湘潭大学 | Multi-unmanned aerial vehicle time window constraint task allocation method based on dynamic proximity threshold communication |
| CN117933669A (en) * | 2024-03-22 | 2024-04-26 | 中国人民解放军国防科技大学 | Dynamic task allocation method and device, computer equipment and storage medium |
| CN117933669B (en) * | 2024-03-22 | 2024-06-21 | 中国人民解放军国防科技大学 | Dynamic task allocation method and device, computer equipment and storage medium |
-
2023
- 2023-05-22 CN CN202310579995.0A patent/CN116610144A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117873137A (en) * | 2024-03-12 | 2024-04-12 | 湘潭大学 | Multi-unmanned aerial vehicle time window constraint task allocation method based on dynamic proximity threshold communication |
| CN117873137B (en) * | 2024-03-12 | 2024-05-17 | 湘潭大学 | Multi-unmanned aerial vehicle time window constraint task allocation method based on dynamic proximity threshold communication |
| CN117933669A (en) * | 2024-03-22 | 2024-04-26 | 中国人民解放军国防科技大学 | Dynamic task allocation method and device, computer equipment and storage medium |
| CN117933669B (en) * | 2024-03-22 | 2024-06-21 | 中国人民解放军国防科技大学 | Dynamic task allocation method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111199360B (en) | Unmanned aerial vehicle task allocation planning method | |
| CN112633654A (en) | Multi-unmanned aerial vehicle task allocation method based on improved cluster expansion consistency bundle algorithm | |
| CN116610144A (en) | Unmanned plane collaborative dynamic task allocation method based on expansion consistency packet algorithm | |
| CN114326827B (en) | Unmanned aerial vehicle cluster multitasking dynamic allocation method and system | |
| CN113671987A (en) | Multi-machine distributed time sequence task allocation method based on non-deadlock contract network algorithm | |
| CN112926827B (en) | Multi-unmanned aerial vehicle task redistribution method based on performance influence algorithm | |
| CN113660681A (en) | Multi-agent resource optimization method applied to unmanned aerial vehicle cluster auxiliary transmission | |
| CN114638155A (en) | Unmanned aerial vehicle task allocation and path planning method based on intelligent airport | |
| CN114371925A (en) | Edge cloud cooperative multi-task scheduling method and system for guaranteeing edge cloud load ratio | |
| Xie et al. | Mathematical problems in engineering improved CNP‐Method‐Based local real‐time cooperative task allocation of heterogeneous multi‐UAV in communication‐constrained environment | |
| CN114189522B (en) | Priority-based blockchain consensus method and system in Internet of vehicles | |
| CN116362327A (en) | Model training method and system and electronic equipment | |
| CN115564374A (en) | Collaborative multitask redistribution method, device, equipment and readable storage medium | |
| CN115115064A (en) | Semi-asynchronous federal learning method and system | |
| CN115689184A (en) | Distributed task allocation method based on consensus binding algorithm | |
| CN116450658A (en) | Node cluster gain-based maximum block storage method and device | |
| CN111782389A (en) | Task scheduling system and method under mobile edge information service network | |
| CN111225380A (en) | Dynamic access method for air-space-earth-sea integrated multi-user cooperative learning | |
| Qian et al. | Online vehicle selection for task replication via bandit learning | |
| CN113176933B (en) | Dynamic cloud network interconnection method for massive workflow tasks | |
| CN117994635B (en) | Federal element learning image recognition method and system with enhanced noise robustness | |
| CN116541831B (en) | Dual defense method based on blockchain and federal learning | |
| CN116909717B (en) | Task scheduling method | |
| CN114675674B (en) | Unmanned aerial vehicle cluster multi-mode heterogeneous value task autonomous cooperative allocation method and system | |
| CN117852745A (en) | Enterprise group distributed decision method, device, equipment and storage medium |
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 |