CN108199868A - A kind of group system distributed control method based on tactics cloud - Google Patents
A kind of group system distributed control method based on tactics cloud Download PDFInfo
- Publication number
- CN108199868A CN108199868A CN201711423867.8A CN201711423867A CN108199868A CN 108199868 A CN108199868 A CN 108199868A CN 201711423867 A CN201711423867 A CN 201711423867A CN 108199868 A CN108199868 A CN 108199868A
- Authority
- CN
- China
- Prior art keywords
- node
- cloud
- tactics
- cost matrix
- matrix
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/12—Discovery or management of network topologies
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Feedback Control In General (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of group system distributed control methods based on tactics cloud, solve the problems such as different optimal in structures carries out slow processing speed during task distribution, shortage overall situation control optimization, information is imperfect.Including:To node structure energy factors a=α × Φ+β τ different in tactics cloud, dissipation factor b=rd2And transmission factorWherein α, β are invariant, and Φ represents the computing capability of node, and τ represents the ability of node-node transmission remote control actuator, and r represents signal transmission obstacle interference coefficient, and d is to need the signal potential difference transmitted in node topology structure;Task object collection is distributed on the node of transmission factor maximum, the cost matrix concentrated on this node to task object is handled, form optimization cost matrix, then the optimization cost matrix is handled using Hungary Algorithm, obtain the handling result of optimization cost matrix, as work distribution chart.
Description
Technical field
The present invention relates to a kind of group system distributed control methods based on tactics cloud, belong to MAS control technology
Field.
Background technology
General multiple agent connection, which makees system, can regard typical group system as.It is each in multiple agent connection makees system
Intelligent body only needs to obtain the information of adjacent intelligent body, you can carries out joint operation.Distributed system has abandoned usual concentration
Controller in this way after Centralized Controller breaks down or Single Controller breaks down, can still carry out whole control, make
The robustness for obtaining total system greatly enhances.Multi-agent system tool has an enormous advantage.Compared with traditional total system,
Distributed system more emphasizes the distributivity in system control process, and the resource that script is concentrated is distributed to each platform, each
On a device, all directions, decentralised control is carried out by controller, you can composition distributed system.
Group system is widely used in the field of military operations.Such as in recent years novel combat mode that Europe, the United States, E Deng states start with
Theories for military operations.Novel theories for military operations are emphasized, large-scale expensive weapon is replaced using small-sized flexible weapon, will make fighting capacity
Amount is broken up on each chain-wales, so as to which war be made a series of flexible variations occur, utilizes more complicated internal coordination and nobody
The consumption of platform/weapon of change can exchange operational superiority for.Compared with traditional operation control system, by using various small-sized forces
The more bulky large-scale weapon of device substitution, the variation for making battlefield is more obvious, and war form is more flexible, battleficld command and control
It is more prone to dynamic and awareness.By the independent learning ability of control system in itself, enable a system to efficiently find
Valid data in mass data are simultaneously calculated accordingly, so as to fulfill making decisions on one's own.
Meanwhile cloud computing due to IT infrastructure using flexible, it is at low cost, service efficiency is high the features such as, civilian
Boundary is widely used.U.S. national defense department has also gradually recognized the advantage and importance of cloud computing technology.It is considered as
The people for coping with increasing information content and ensureing to need can timely use a kind of important means of information.U.S. Department of Defense is building
Cloud computing is clearly just classified as one of key technology, the information system of U.S. Department of Defense when its if " united information environment " (JIE)
Defense Information Systems Agency of authorities (DISA) also constantly widelys popularize its enterprise-level cloud computing basis in entire Ministry of National Defence's range
Facility " military cloud " (MilCloud).And so-called " tactics cloud ", it seeks to cloud computing extending to tactical environment.Tactics cloud allows letter
Cease treatment facility from combatant closer to, combatant is allowed in processing locality data and historical data to be retrieved, so as to shorten
Information is obtained to the time using information, realizes a degree of autonomous fight capability.Tactics cloud can also enhance combatant's
Ability provides tactics network edge typically no function.
Currently the frame of process problem has been formed on the basis of centralized business cloud.Exist including Hadoop and spark
Interior frame uses in enterprise, indicates the maturation of business cloud computing.However, in battlefield, environmental resource by
The features such as limit, data volume are big, bad environments, low and unstable communication bandwidth and mobile equipment size, processing capacity, storage energy
Power, battery capacity are limited, and cause tactics cloud dynamic extremely strong, and the uncertainty for calculating control operational node is extremely strong.Currently for
The group system distributed command control method of tactics cloud is also in the exploratory stage.
Invention content
The present invention provides a kind of group system distributed control methods based on tactics cloud, and what is mainly solved is novel
In combat mode, combat duty links ability breaks up multiple and different platform weapons, using different optimal in structures into
The problems such as processing speed that row task generates when distributing is slowly, the shortage overall situation controls optimization, information is imperfect.
The present invention is achieved through the following technical solutions:
A kind of group system distributed control method based on tactics cloud, including:
To node structure energy factors a=α × Φ+β τ each in tactics cloud, dissipation factor b=rd2And transmission factorWherein α, β are invariant, and Φ represents the computing capability of node, and τ represents the ability of node-node transmission remote control actuator, r
Represent signal transmission obstacle interference coefficient, d is to need the signal potential difference transmitted in node topology structure;
Task object collection is distributed on the node of transmission factor maximum, the cost concentrated on this node to task object
Matrix is handled, and forms optimization cost matrix, and then the optimization cost matrix is handled using Hungary Algorithm, is obtained
To the handling result Ans of optimization cost matrix, as work distribution chart, distributed command control is completed according to the work distribution chart
System.
Further, the task object collection is handled in the following ways:First by the real-time calculating task stream of sensor with
One real time input data period is duration, then to real-time streams according to other side's optimal in structure different types of in the period into line number
It is encapsulated according to decomposing, forms single task role object set.
Further, the optimization cost matrix is formed in the following ways:In node in single task role object set
Comprising cost matrix A processing, build specific cost matrix H so that Hij=AijThe line number of-k, wherein i representing matrixes, j
The row number of representing matrix;When the dimension of H is excessively high, it is 4 that H is resolved into number by iterative processingn, H ∈ RdMatrix in block form, whereinSo that each matrix decomposition is no more than 10 square formation for dimension, the matrix in block form after decomposition is final optimization cost
Matrix.
Further, after obtaining Ans, judge whether there is reproducible results in Ans, if so, then being screened to Ans, structure
Cost matrix B is built, Hungary Algorithm is then reused to B matrixes and carries out matrix disposal, until not containing repetition knot in Ans
Fruit, so as to obtain final work distribution chart.
Beneficial effects of the present invention:
When the 1st, distributing task the present invention is based on tactics cloud, subtask distribution is carried out using transmission factor k, optimizes tactics cloud
The Map processes of the specific big data problem of child node processing, have preferably handled process problem in de-centralized dynamic tactics cloud
Computational efficiency.
2nd, original cost matrix is carried out piecemeal dimensionality reduction by the present invention, be ensure that the low-dimensional number of single cost matrix, is made algorithm
Complexity by original Hungary Algorithm n3It is transformed into n3/4.It ensure that Task Allocation Problem can be in the real-time streams single treatment period
Middle completion adapts to battlefield and responds rapidly to background.
3rd, the present invention distributes task to traditional Hungary Algorithm and is combined with tactics cloud node calculation processing ability, by sensor
Capacity of equipment collectively forms processing target in the enemy's information and tactics cloud that detect.It ensure that the information completely for calculating distribution
The robustness of property, standard and system.
Description of the drawings
Fig. 1-the present invention is based on the group system distributed control method schematic diagrames of tactics cloud;
Fig. 2-node-node transmission of the present invention factor schematic diagram;
Fig. 3-the method for the present invention specific implementation flow chart.
Specific embodiment
The invention will be described further below in conjunction with the accompanying drawings.
In order to verify the feasibility for the group system distributed control method for being based on " tactics cloud ", We conducted emulation to grind
Study carefully.The present invention will be further described with example below in conjunction with the accompanying drawings:
Step 1: task reconfiguration
As shown in Figure 1, carrying out the control of group system using " tactics cloud ", need to calculate spacial ability with entire high in the clouds
Composition and control based on.Optimal in structures all in battlefield are realized into information communication, form de-centralized tactics cloud, simultaneously will
Battlefield operating personnel's handheld terminal introduces cloud, forms entire tactics cloud.It carries out using during node distribution in the present invention, inside cloud
Hadoop and Spark design frameworks.Distributed data processing process in cloud uses the MapReduce thoughts of big data processing.
After being decoupled by problem, data are subjected to distributed storage with arranging, carry out the operation of data in entire cluster later, it is complete
It is integrated after into calculating, so as to complete to calculate body tasks.Make data dynamic rapid mobile between the individual nodes, so as to make
Using data calculating nearby rather than program operation nearby when must handle.Because required for mobile data ratio is issued program
Time it is shorter, it is more efficient.
Due to battlefield form complexity, the event that each second occurs is not expectable, causes the needing to distribute of the task extremely complex.
In order to solve very big challenge, the first step of the present invention decomposes the coupled problem:With single enemy's optimal in structure pair
Object has the operational weapon distribution task of one's own side as arriving single object, by identical in one in multiple real-time streams of time
Type weapon is transmitted the calculation processing platform in tactics cloud as single task role object set.
Because being contained in real-time stream by the collected enemy's information for needing to handle distribution task of sensor, beating
It needs to pre-process data during packet, obtains the cost matrix A for enemy's platform specific.It, will entire battlefield after obtaining A
Task Allocation Problem reconstructed, form a series of fractionation problems by object of each enemy's platform specific.Entirely
Operation problem is converted into problems with:Which kind of warfare equipment which kind of combat duty distribution carried out using for single enemy's platform.
Step 2: data distribution
When single task role object set being sent to calculate node being handled, due to the processing of each node in tactics cloud
Computing capability is different, and the flow for transmitting data is also different, therefore defines war using energy factors and dissipation factor in the present invention
The processing capacity of art cloud single node, optimizes the distribution procedure of single task role object set.
As shown in Fig. 2, energy factors are related to the processing capacity of node.Processing capacity includes the computing capability of node and appoints
Business fight capability.Computing capability and CPU, disk read-write etc. correlations;Task fight capability is performed with the direct contactless transmission of node
The speed of device is related.
Energy factors a=α * Φ+β τ are defined, wherein α, β is invariant, and Φ represents the computing capability of node, and τ represents section
The ability of point transmission remote control actuator;
Define dissipation factor b=rd2, for r to represent signal transmission obstacle interference coefficient, d is to be needed in node topology structure
The signal potential difference of transmission.
Transmission factor k=a/b is defined, as the leading indicator in cloud node data transmission process.
In data transmission procedure, single task role object set choose the transmission of giant's connecting path in specific topological structure because
Sub maximum node carries out data transmission.Data distribution after optimization, each calculate node only need to calculate a small amount of number
According to, at the same by the cost optimization of transmission to minimum.Therefore calculate node processing capacity greatly increases.
The process of data distribution has relied on the Distributed Calculation of tactics cloud and the ability of parallel computation.Distributed Calculation ensures
Meet same task between multiple CPU and multiple memories jointly between each physical node of cloud server cluster;Parallel meter
Calculation meets same task jointly after ensure that Cloud Server virtualization between single CPU and multiple memories.By data distribution mistake
Journey ensure that the comprehensive utilization of " high in the clouds " calculated performance.
Step 3: build specific cost matrix and optimization cost matrix
Cost matrix A processing to being included in single task role object set in node, builds specific cost matrix H,
Cause Hij=AijThe line number of-k, wherein i representing matrixes, the row number of j representing matrixes;AijRepresent each element in A matrixes,
Here it is to be represented each element in A subtracting the k factors with mathematical form, H ∈ R hered, represent H in d linear nothings
It closes in the d dimension spaces that vector is turned into, when the dimension of H is excessively high, it is 4 that H is resolved into number by iterative processingn, H ∈ RdPiecemeal square
Battle array, whereinSo that each matrix decomposition is no more than 10 square formation for dimension, the matrix in block form after decomposition is final
Optimize cost matrix.
Since the complexity of Hungary Algorithm original in task assignment procedure is n3, when the dimension of cost matrix is excessively high,
The overlong time needed for processing procedure in single node, it is more than the task reconfiguration period to have maximum probability, reduces whole system
Robustness.In order to optimize calculating solution procedure, task distribution Hungary Algorithm is optimized:When the dimension of H is excessively high, iteration
It is 4 that H is carried out processing to resolve into number by processingnMatrix in block form so that each matrix in block form is no more than 10 square formation for dimension.
Matrix in block form is to optimize cost matrix.After structure optimization cost matrix, the first step complexity for the task of distributing is reduced to (n/2
)3, greatly reduce the complexity that single node processing solves.Ensure that solution procedure is completed within the task reconfiguration period.
Step 4: processing array must solve
The present invention solves distributed command and the control of group system using tactics cloud.It, will for the commanding and decision-making in battlefield
The overall scheduling in battlefield moves on to high in the clouds with optimization distribution task.By reasonable distribution task so that complete the cost of required by task
It is minimum.
Overall minimum Task Allocation Problem can use n2A decision variable description.Enable xijFor 0-1 decision variables, i.e. xij=
1 represents that i-th of role is arranged to perform jth item task, xij=0 represents that i-th of role is not arranged to go to complete jth item task, uses cij
Represent that i-th of role completes the cost (real number) of jth item required by task, then cijCost matrix known to composition.Then:
xij=0or1, i, j ∈ N=1,2 ... n } (1)
In overall linear planning problem.The matrix disposal method that we employ traditional Hungary Algorithm carries out overall wire
Property planing method solve.The core of Hungary Algorithm is exactly to find augmenting path, it is a kind of to seek bipartite graph most with augmenting path
Big matched algorithm.Its principle is as shown in Figure 3.
For overall linear planning problem existential theorem 1:From (Cij) matrix often row (each column) subtract or plus one often
Number ui(or vj) form new matrix (C 'ij),C′ij=Cij±ui+vj, then corresponding (C 'ij) xijOptimal solution and original (Cij) most
Excellent solution is of equal value.I.e.:
The detailed process of Hungary Algorithm Matrix Solving part is as follows:
(1) all numbers that often goes subtract the minterm of the row.
(2) all numbers of each columns subtract the minterm of the row.
(3) uses horizontal line or vertical line across all 0 in matrix, and records the minimum circuit reached needed for this purpose
Sum.
(4) if circuits sum be equal to matrix line number or columns n, then a kind of optimal distribution be it is possible, it is complete
Into.If sum is less than n, next step is performed.
(5) finds the minterm in the unlapped place of circuit, and the row there are unlapped item subtracts this, then should
Item is added in the row of covering.
Hungary Algorithm is applicable not only to the square formation that object function is minimal type.When assignment problem target call is very big,
AskAt this point, large can will be asked, which to be changed into, seeks minimal type.Utilize formula:MaxZ=
Min(-Z)
I.e.:
Again by theorem 1, matrix-CijThe often row of n*n can add constant M and (or enable CijMiddle greatest member is for M
Can), enable bij=M-Cij, at this moment coefficient matrix can be transformed into:B=((bij)nm, at this moment, bij>=0, B are the condensation matrix of C.
When coefficient matrix is not square formation, target is still Min type problems.Solving at this time needs to add coefficient by cost matrix
It is converted into square formation.
The optimization cost matrix in step 4 is handled using matrix disposal step in traditional Hungary Algorithm, can be obtained
To specific task allocation result.The algorithm can simultaneously be run in each calculate node, handle the data of distribution, form task point
With table.
Step 5: data fusion
Handling result after single leg gusset is calculated, due to step 4 optimization cost matrix structure,
Lead to the result of calculation that there may be multiple tasks in handling result for single enemy's optimal in structure.In the present invention, to multiple
Result of calculation carries out data fusion:Screening cost matrix B is built to allocation result, the dimension of screening cost matrix B is right less than 10
Screening cost matrix B is handled using process shown in step 4, and final times after reproducible results is merged is obtained after calculating
Business allocation table.
After result of calculation has been handled, the control bit of node free time restores, which comes back to idle mode, waits for number
According to departure process, step 2 is returned to.
It is complete to be beamed back execution optimal in structure by data processing and calculating process by entire high in the clouds controller for result of calculation
Into task, entire complete system is formed.
In emulation experiment, it is based on tactics cloud using the method for the present invention and distributed command control is carried out to group system, prevent
It is imperial target occur suddenly.After emulation, the infrastructure device defence success rate of one's own side is 100%.It can be obtained by simulation result
Conclusion, the distributed command control method of the large-scale cluster system based on tactics cloud ensure that the reasonable utilization of global resource,
With high superiority.
Above-described is only presently preferred embodiments of the present invention, and the present invention is not limited solely to above-described embodiment, all
The local change done within the spirit and principles in the present invention, equivalent replacement, improvement etc. should be included in protection of the invention
Within the scope of.
Claims (4)
1. a kind of group system distributed control method based on tactics cloud, which is characterized in that including:
To node structure energy factors a=α × Φ+β τ each in tactics cloud, dissipation factor b=rd2And transmission factorIts
Middle α, β are invariant, and Φ represents the computing capability of node, and τ represents the ability of node-node transmission remote control actuator, and r represents signal
Obstacle interference coefficient is transmitted, d is to need the signal potential difference transmitted in node topology structure;
Task object collection is distributed on the node of transmission factor maximum, the cost matrix concentrated on this node to task object
It is handled, forms optimization cost matrix, then the optimization cost matrix is handled using Hungary Algorithm, is obtained excellent
Change the handling result Ans of cost matrix, as work distribution chart, distributed command control is completed according to the work distribution chart.
2. a kind of group system distributed control method based on tactics cloud as described in claim 1, which is characterized in that described
Task object collection is distributed to after handling in the following ways on the node of transmission factor maximum again:First by the real-time calculating of sensor
Then task flow draws to real-time streams according to other side different types of in the period using a real time input data period as duration
Platform carries out data and decomposes encapsulation, forms single task role object set.
3. a kind of group system distributed control method based on tactics cloud as claimed in claim 2, which is characterized in that described
Optimization cost matrix is formed in the following ways:In node at the cost matrix A that is included in single task role object set
Reason, builds specific cost matrix H so that Hij=AijThe line number of-k, wherein i representing matrixes, the row number of j representing matrixes;When H's
When dimension is excessively high, it is 4 that H is resolved into number by iterative processingn, H ∈ RdMatrix in block form, whereinSo that each matrix point
Solve for dimension be no more than 10 square formation, the matrix in block form after decomposition is final optimization cost matrix.
4. a kind of group system distributed control method based on tactics cloud as described in claims 1 or 2 or 3, feature exist
In after obtaining Ans, judging whether there is reproducible results in Ans, if so, then screened to Ans, build cost matrix B, so
Hungary Algorithm is reused to B matrixes afterwards and carries out matrix disposal, it is final so as to obtain until not containing reproducible results in Ans
Work distribution chart.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711423867.8A CN108199868B (en) | 2017-12-25 | 2017-12-25 | Distributed control method for cluster system based on tactical cloud |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711423867.8A CN108199868B (en) | 2017-12-25 | 2017-12-25 | Distributed control method for cluster system based on tactical cloud |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108199868A true CN108199868A (en) | 2018-06-22 |
CN108199868B CN108199868B (en) | 2020-12-15 |
Family
ID=62583790
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711423867.8A Active CN108199868B (en) | 2017-12-25 | 2017-12-25 | Distributed control method for cluster system based on tactical cloud |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108199868B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110955734A (en) * | 2020-02-13 | 2020-04-03 | 北京一流科技有限公司 | Distributed signature decision system and method for logic node |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102448123A (en) * | 2012-01-16 | 2012-05-09 | 河海大学常州校区 | Task allocation algorithm in wireless sensor network based on node property |
CN103595805A (en) * | 2013-11-22 | 2014-02-19 | 浪潮电子信息产业股份有限公司 | Data placement method based on distributed cluster |
US20140148175A1 (en) * | 2012-11-28 | 2014-05-29 | China Mobile Communications Corporation | Load balancing method and device |
CN104539551A (en) * | 2014-12-22 | 2015-04-22 | 清华大学 | Virtual service migration method for routing and switching platform |
CN106126315A (en) * | 2016-06-17 | 2016-11-16 | 广东工业大学 | A kind of virtual machine distribution method in the data center of minimization communication delay |
CN106886459A (en) * | 2017-01-24 | 2017-06-23 | 浙江工商大学 | A kind of multiple agent internet data acquisition tasks distribution method based on actual measurement bandwidth |
CN106919389A (en) * | 2017-02-24 | 2017-07-04 | 湖北大学 | Based on the software development resource automatic scheduling method and system that improve Hungary Algorithm |
-
2017
- 2017-12-25 CN CN201711423867.8A patent/CN108199868B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102448123A (en) * | 2012-01-16 | 2012-05-09 | 河海大学常州校区 | Task allocation algorithm in wireless sensor network based on node property |
US20140148175A1 (en) * | 2012-11-28 | 2014-05-29 | China Mobile Communications Corporation | Load balancing method and device |
CN103595805A (en) * | 2013-11-22 | 2014-02-19 | 浪潮电子信息产业股份有限公司 | Data placement method based on distributed cluster |
CN104539551A (en) * | 2014-12-22 | 2015-04-22 | 清华大学 | Virtual service migration method for routing and switching platform |
CN106126315A (en) * | 2016-06-17 | 2016-11-16 | 广东工业大学 | A kind of virtual machine distribution method in the data center of minimization communication delay |
CN106886459A (en) * | 2017-01-24 | 2017-06-23 | 浙江工商大学 | A kind of multiple agent internet data acquisition tasks distribution method based on actual measurement bandwidth |
CN106919389A (en) * | 2017-02-24 | 2017-07-04 | 湖北大学 | Based on the software development resource automatic scheduling method and system that improve Hungary Algorithm |
Non-Patent Citations (3)
Title |
---|
展红英: "相控阵雷达系统实时任务负载分配仿真研究", 《中国优秀硕士学位论文全文数据库(电子期刊)》 * |
董帅君: "行动规划中的任务排序和目标分配研究", 《中国优秀硕士学位论文全文数据库(电子期刊)工程科技II辑》 * |
韩攀: "多无人机协同任务规划技术", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110955734A (en) * | 2020-02-13 | 2020-04-03 | 北京一流科技有限公司 | Distributed signature decision system and method for logic node |
Also Published As
Publication number | Publication date |
---|---|
CN108199868B (en) | 2020-12-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Rosalie et al. | Chaos-enhanced mobility models for multilevel swarms of UAVs | |
CN101382968B (en) | Advanced synthesis type aviation electronic simulation system and simulation method thereof | |
CN106776796B (en) | Unmanned aerial vehicle task planning system and method based on cloud computing and big data | |
CN105184092B (en) | Polymorphic type unmanned plane cotasking distribution method under the conditions of a kind of resource constraint | |
Zhao et al. | An approach for weapon system-of-systems scheme generation based on a supernetwork granular analysis | |
CN101819661A (en) | Method for analyzing and evaluating decision supporting capability of complex system | |
CN111460621A (en) | Modularized simulation modeling method and system | |
CN112766595B (en) | Command control device, method, system, computer equipment and medium | |
CN110348070A (en) | A kind of system modeling method of model-based system engineering and super-network theory | |
Cai et al. | Theory and application of multi-robot service-oriented architecture | |
CN106815471A (en) | A kind of special vehicle information system efficiency estimation method | |
CN107220168B (en) | Information system maturity evaluation method based on integration | |
CN112884164B (en) | Federal machine learning migration method and system for intelligent mobile terminal | |
CN115276753A (en) | Low-orbit broadband satellite constellation compatible simulation system | |
CN108199868A (en) | A kind of group system distributed control method based on tactics cloud | |
Yang et al. | Learning graph-enhanced commander-executor for multi-agent navigation | |
CN116700340A (en) | Track planning method and device and unmanned aerial vehicle cluster | |
Gaowei et al. | Using multi-layer coding genetic algorithm to solve time-critical task assignment of heterogeneous UAV teaming | |
Skobelev et al. | Using multi-agent technology for the distributed management of a cluster of remote sensing satellites | |
CN116166225A (en) | Task planning middle stage frame design method based on software definition | |
Wang et al. | OPM & color petri nets based executable system of systems architecting: A building block in FILA-SoS | |
CN106851689B (en) | Multifactor method for allocating tasks in wireless sensor and actor network | |
Tuba et al. | Water cycle algorithm for robot path planning | |
CN114881526A (en) | Comprehensive work task efficiency evaluation method for multi-platform avionics system | |
Airlangga et al. | A novel architectural design for solving lost-link problems in UAV collaboration |
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 |