CN110688754A - Combat system architecture modeling and optimal search method - Google Patents

Combat system architecture modeling and optimal search method Download PDF

Info

Publication number
CN110688754A
CN110688754A CN201910912549.0A CN201910912549A CN110688754A CN 110688754 A CN110688754 A CN 110688754A CN 201910912549 A CN201910912549 A CN 201910912549A CN 110688754 A CN110688754 A CN 110688754A
Authority
CN
China
Prior art keywords
network
task
nodes
ososa
combat
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
Application number
CN201910912549.0A
Other languages
Chinese (zh)
Inventor
王维平
周鑫
王涛
朱一凡
李小波
井田
李童心
段婷
王彦锋
黄美根
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
National University of Defense Technology
Original Assignee
National University of Defense Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by National University of Defense Technology filed Critical National University of Defense Technology
Priority to CN201910912549.0A priority Critical patent/CN110688754A/en
Publication of CN110688754A publication Critical patent/CN110688754A/en
Priority to PCT/CN2020/109335 priority patent/WO2021057329A1/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/18—Network design, e.g. design based on topological or interconnect aspects of utility systems, piping, heating ventilation air conditioning [HVAC] or cabling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation

Abstract

The invention discloses a combat system architecture modeling and optimal searching method which comprises an OSoSA formal definition and an OSoSA search based on a super network, wherein the OSoSA is composed of three heterogeneous networks, namely a task network, a system network and a command network, the task network comprises task nodes, the system network comprises system nodes, the command network comprises command nodes, the combat capability of the OSoSA formal definition is obtained by the OSoS, and the task network, the system network and the command network jointly form a combat system. In the application process, the average return value can be in a higher state compared with the traditional method, and the method is further favorable for popularization and application.

Description

Combat system architecture modeling and optimal search method
Technical Field
The invention relates to the technical field of combat systems, in particular to a method for modeling architecture and optimally searching for a combat system.
Background
Many scholars model the architecture of the combat system into a super-network model, and some researchers provide a military communication super-network structure taking a network as a center. The network consists of five heterogeneous nodes: the system comprises sensor nodes, information nodes, decision nodes, communication nodes and effect nodes. Researchers have proposed a super-network model of a command control system, which includes observation nodes, command control nodes, effect nodes, and relationships among three types of nodes. In addition, researchers establish a weapon equipment system hyper-network model based on a network center mode, and provide a granularity analysis method for reducing complexity of a weapon system generation scheme based on a quotient space theory. Researchers have also proposed a multi-layer command control hyper-network model based on attribute collaborative priority and hyper-graph theory. These studies were conducted separately from the viewpoint of equipment systems, functions, and command and control structures, and no consideration was given to the system capacity generation elements, and therefore, these research methods and models had low average return values after actual experiments. Therefore, a combat system architecture modeling and optimal search method is provided.
Disclosure of Invention
The invention aims to solve the defects in the prior art and provides a combat system architecture modeling and optimal search method.
In order to achieve the purpose, the invention adopts the following technical scheme:
a combat system architecture modeling and optimal search method comprises an OSoSA formal definition and an OSoSA search based on a super network, wherein the OSoSA formal definition comprises three heterogeneous networks, namely a task network, a system network and a command control network, the task network comprises task nodes, the system network comprises system nodes, the command control network comprises command control nodes, the combat capability of the OSoSA formal definition is obtained by the OSoS, and the task network, the system network and the command control network jointly form a combat system.
Preferably, the task nodes of the task network are combat activities, denoted T, that can be performed by the equipment system.
Preferably, the system node of the task network refers to equipment having a specific function and capable of completing a specific task, and is denoted as S.
Preferably, the instruction nodes of the instruction network are logic nodes for processing information, managing organization, decision planning, controlling feedback, and the like, and are denoted as C.
The invention provides a combat system architecture modeling and optimal search method, which has the beneficial effects that: in the application process, the average return value of the scheme can be in a higher state compared with that of a traditional method, and therefore the scheme is favorable for popularization and application.
Drawings
FIG. 1 is an exemplary diagram of a task network of the present invention;
FIG. 2 is an exemplary diagram of a system network of the present invention;
FIG. 3 is a bipartite graph illustrating the correspondence between a task network and a system network according to the present invention;
FIG. 4 is an exemplary diagram of an instructional network of the present invention;
FIG. 5 is a bipartite graph of the correspondence between the system network and the command network according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.
In the description of the present invention, it is to be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present invention.
Referring to fig. 1-5, a combat system architecture modeling and optimal search method comprises an OSoSA formal definition and an OSoSA search based on a super network, wherein the OSoSA formal definition comprises three heterogeneous networks, namely a task network, a system network and a command network, the task network comprises task nodes, the system network comprises system nodes, the command network comprises command nodes, the combat capability of the OSoSA formal definition is obtained by the OSoS, and the task network, the system network and the command network jointly form a combat system.
The task nodes of the task network are combat activities that can be executed by the equipment system, denoted T, and in order to complete a system mission, the mission should be broken down into a series of executable tasks, i.e. task links; the task links can be abstracted into a directed graph, a system mission can be decomposed into different task links, each task link is provided with a starting task node and an ending task node, different task links can have different efficiencies, the task links mainly have two types of logical relations, namely sequential execution and parallel execution, and a plurality of missions correspond to a plurality of task links forming a task network and are marked as GT＝(VT,ET) And as shown in fig. 1.
The system node of the task network refers to equipment with specific functions and capable of completing specific tasks, denoted as S, in the task planning method, an equipment system with specific functions is used for completing specific tasks, and thus the relationship between system nodes (such as unmanned aerial vehicles, tanks and naval vessels) is influenced by the task node, as shown in fig. 2: the system network is denoted GS＝(VS,ES) (ii) a The correspondence between task nodes and system nodes can be defined as a bipartite graph, as in fig. 3: is marked as GTS＝(VTS,ETS)。
The instruction control nodes of the instruction control network are used for processing logical nodes of information, management organization, decision planning, control feedback and the like, and are represented as C, the instruction control nodes process instruction information of a superior level and an inferior level and keep information interaction with the instruction control nodes of a same level, so that the instruction control network has an organization structure formed by instruction relations of all the instruction control nodes, and the organization structure is represented as GC＝(VC,EC) As in fig. 4; most organization structures are modeled as, but not limited to, a tree graph, and in addition, the corresponding relation between the system nodes and the control nodes is defined as a bipartite graph marked as GSC＝(VSC,ESC) As in fig. 5.
Since the combat capability defined formally by an OSoSA is derived by an OSoS, the OSoS may be affected by factors other than OSoSA, which may lead to a surge. Thus, the OSoS capabilities based on OSoSA may not be unique, and thus each OSoSA scheme has an uncertain potential reward value.
The commander needs to select one architecture scheme (simply referred to as scheme) in the architecture scheme space to develop the combat system. For simplicity, this document models the commander, decision-making and consulting authorities as agents. The OSoS capability is measured by a reward value. Reward value x for each schemekSubject to the probability distribution fk (xk), the return values of the different schemes are independent of each other. Where K ∈ K, K ∈ {1,2, …, | K | }, K being the number of solutions in the solution space. The return value for each architecture is uncertain in advance, but can be obtained by developing OSoS or consulting other agents. Agents continuously explore the schemes in the undeveloped scheme space and finally select one scheme from all the developed scheme spaces as a final option. The goal of the Agent is to select an architecture with the highest expected return value and the least accumulated search cost.
Wherein the unknown state indicates that the scheme has not been developed and its reported value is unknown; the known state indicates that the scheme has been developed and that its return value is known. The search state represents the reward value for which the solution is being queried. Before exploring a scenario, the scenario has potential reward values. After developing an OSoS, a return to the solutionThe reported value is known. The actions of the Agent include: self-developed, developed by other agents, and consulted related agents. In particular, based on having a cost of
Figure BDA0002215158600000051
After the scheme k of (a) is developed, the unknown state is converted to a known state. In addition, an Agent may request other agents to
Figure BDA0002215158600000052
Cost of developing the OSoS. In addition, an Agent may consult related agents, such as an organization or department, that may have completed similar work. Cost of the consultation procedure is noted
Example 1 the invention is described by taking scheme k as an example, and four binary decision variables are defined as follows: if the scheme is developed by the Agent itself, thenOtherwise
Figure BDA0002215158600000055
If the scheme is developed by other Agents, then
Figure BDA0002215158600000056
Otherwise
Figure BDA0002215158600000057
If an Agent chooses to consult the relevant Agent, then
Figure BDA0002215158600000058
Otherwise
Figure BDA0002215158600000059
If scheme k is selected as the final scheme, then sk1, otherwise sk0. The OSoSA search problem (OSoSOSP) framework is as follows:
CSoSAS:
S.t.
Figure BDA0002215158600000062
Figure BDA0002215158600000063
Figure BDA0002215158600000064
Figure BDA0002215158600000065
Figure BDA0002215158600000066
Figure BDA0002215158600000067
Figure BDA0002215158600000068
the objective function is the sum of maximizing the reward value of the developed architecture and minimizing the cumulative search cost. Specifically, constraint (a) ensures that either solution has been either developed or not developed. Constraint (b) indicates that if a solution is finally selected, it must have been developed. Constraint (c) means that only one scheme is finally selected. Constraint (d) represents the value space of the four decision variables. Constraint (e) represents the number of times an Agent requests the relevant Agent. Constraint (f) refers to the discount rate, representing the impact of development time on the return value. Constraint (g) represents the cost per action.
Further solve OSoSA dynamic programming problem.
(1) Decision index
Each plan k defines an execution action
Figure BDA0002215158600000069
And
Figure BDA00022151586000000610
wherein the indexes are respectively recorded as
Figure BDA00022151586000000611
And
Figure BDA00022151586000000612
Figure BDA00022151586000000613
Figure BDA00022151586000000614
further, in the present invention,
Figure BDA0002215158600000071
Figure BDA0002215158600000072
Figure BDA0002215158600000073
according to the stateAnd index set
Figure BDA0002215158600000075
Designing a simple but optimal search rule, scoreTo judge rules and select rules. The judgment rule means that if the Agent needs to further explore the architecture with unknown effect, an unknown architecture with the maximum index is selected. Meanwhile, actions are selected according to the maximum indexes, namely self research and development, company development and help seeking for relevant mechanisms. The stopping rule refers to stopping searching and selecting the architecture with the maximum return value as a scheme when the maximum collected sampling return value is larger than the development index, the development index and the consultation index of all the location architectures.
(2) Search algorithm
The calculation of each index is independent and is not influenced by the probability distribution of the return value of other schemes. The GSDP is composed of an index calculation program, a sequential search program, and a framework development program. Specifically, the Agent first calculates the decision indexes of all schemes according to a formula. Secondly, the indexes are sorted according to a sorting method, such as a heap sorting method, and sorting results are stored in a vector pi. Third, the sequence searching procedure is performed to obtain the best solution.
Figure BDA0002215158600000081
In the sequence searching program, an optimal architecture scheme can be calculated through K iterations at most. And comparing the current maximum sampling value y with the maximum index pi (0) in each iteration according to a set rule. If the maximum sampling value is not less than the maximum index, stopping the search, and taking the architecture m with the current maximum sampling return value as the selected architecture. Otherwise, Executing the Executing program to continue searching according to the architecture index i corresponding to pi (0) and the action a. Updating the variable if the sampled reward value for architecture i is obtained
Figure BDA0002215158600000082
Wherein
Figure BDA0002215158600000083
Representing a removed set
Figure BDA0002215158600000084
Architecture i in (1).
Figure BDA0002215158600000091
In the Executing program, if the action taken is consultation, it is determined whether a report value of the architecture i can be obtained through a relevant mechanism, that is, it is determined whether the Available is true. Wherein "-" denotes sampling, yi～Fi(xi) Represents a pair probability distribution Fi(xi) Sampling is performed.
Figure BDA0002215158600000101
Example 1 is applied to a scheme space scene of 100 examples, and the average return value of the scheme space scene at least exceeds 17.6% of the average return value of the optimal algorithm; in a scheme space scene with the scheme number of 10000, the average return value at least exceeds 15.2% of the average return value of the optimal algorithm; in a scheme space scene with the number of schemes of 1000000, the average return value of the scheme space scene at least exceeds 21.9% of the average return value of the optimal algorithm, and the average return value in the invention can be in a higher state by combining the data of the average return value of the optimal algorithm, so that the invention is favorable for popularization and application.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.

Claims (4)

1. The combat system architecture modeling and optimal search method is characterized by comprising an OSoSA formal definition and an OSoSA search based on a super network, wherein the OSoSA is composed of three heterogeneous networks, namely a task network, a system network and a command network, the task network comprises task nodes, the system network comprises system nodes, the command network comprises command nodes, the combat capability of the OSoSA formal definition is obtained by the OSoS, and the task network, the system network and the command network jointly form a combat system.
2. The method of claim 1, wherein the task nodes of the task network are combat activities executable by equipment systems and denoted as T.
3. The tactical architecture modeling and optimal search method of claim 1, wherein the system node of the task network is a device with specific function and capable of completing a specific task, denoted as S.
4. The tactical architecture modeling and optimal search method of claim 1, wherein the commanding nodes of said commanding network are logical nodes for processing information, management organization, decision planning, control feedback, etc., denoted C.
CN201910912549.0A 2019-09-25 2019-09-25 Combat system architecture modeling and optimal search method Pending CN110688754A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201910912549.0A CN110688754A (en) 2019-09-25 2019-09-25 Combat system architecture modeling and optimal search method
PCT/CN2020/109335 WO2021057329A1 (en) 2019-09-25 2020-08-14 Combat system architecture modeling and optimal searching method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910912549.0A CN110688754A (en) 2019-09-25 2019-09-25 Combat system architecture modeling and optimal search method

Publications (1)

Publication Number Publication Date
CN110688754A true CN110688754A (en) 2020-01-14

Family

ID=69110265

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910912549.0A Pending CN110688754A (en) 2019-09-25 2019-09-25 Combat system architecture modeling and optimal search method

Country Status (2)

Country Link
CN (1) CN110688754A (en)
WO (1) WO2021057329A1 (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111967741A (en) * 2020-08-04 2020-11-20 中国人民解放军国防科技大学 EC 2-based cloud fluidization command architecture design method for unmanned combat system
CN112422353A (en) * 2021-01-25 2021-02-26 中国人民解放军国防科技大学 Military force distribution network generation method based on effectiveness
WO2021057329A1 (en) * 2019-09-25 2021-04-01 中国人民解放军国防科技大学 Combat system architecture modeling and optimal searching method
CN112632744A (en) * 2020-11-13 2021-04-09 中国人民解放军国防科技大学 Combat system architecture modeling method and space exploration algorithm based on hyper-network model
CN112801539A (en) * 2021-02-23 2021-05-14 中国人民解放军国防科技大学 Flexible network architecture dynamic scheduling model of unmanned aerial vehicle cluster task
CN114154322A (en) * 2021-11-29 2022-03-08 上海烜翊科技有限公司 System overall design method output by system architecture model
GB2618302A (en) * 2020-11-13 2023-11-08 National Univ Of Defense Technology Hypernetwork model-based organization architecture modeling method and space exploration algorithm

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9208335B2 (en) * 2013-09-17 2015-12-08 Auburn University Space-time separated and jointly evolving relationship-based network access and data protection system
CN106603309B (en) * 2017-01-04 2019-04-19 大连大学 A kind of charge network hierarchy evolution method based on super-network
CN110688754A (en) * 2019-09-25 2020-01-14 中国人民解放军国防科技大学 Combat system architecture modeling and optimal search method

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021057329A1 (en) * 2019-09-25 2021-04-01 中国人民解放军国防科技大学 Combat system architecture modeling and optimal searching method
CN111967741A (en) * 2020-08-04 2020-11-20 中国人民解放军国防科技大学 EC 2-based cloud fluidization command architecture design method for unmanned combat system
CN112632744A (en) * 2020-11-13 2021-04-09 中国人民解放军国防科技大学 Combat system architecture modeling method and space exploration algorithm based on hyper-network model
WO2022099915A1 (en) * 2020-11-13 2022-05-19 中国人民解放军国防科技大学 Hypernetwork model-based organization architecture modeling method and space exploration algorithm
GB2618302A (en) * 2020-11-13 2023-11-08 National Univ Of Defense Technology Hypernetwork model-based organization architecture modeling method and space exploration algorithm
CN112422353A (en) * 2021-01-25 2021-02-26 中国人民解放军国防科技大学 Military force distribution network generation method based on effectiveness
CN112801539A (en) * 2021-02-23 2021-05-14 中国人民解放军国防科技大学 Flexible network architecture dynamic scheduling model of unmanned aerial vehicle cluster task
CN114154322A (en) * 2021-11-29 2022-03-08 上海烜翊科技有限公司 System overall design method output by system architecture model

Also Published As

Publication number Publication date
WO2021057329A1 (en) 2021-04-01

Similar Documents

Publication Publication Date Title
CN110688754A (en) Combat system architecture modeling and optimal search method
CN112632744B (en) Combat architecture modeling method and space exploration method based on super network model
Ferguson et al. Anytime, dynamic planning in high-dimensional search spaces
CN105467997A (en) Storage robot path program method based on linear temporal logic theory
Sun et al. An artificial bee colony algorithm with random location updating
Lee et al. Monte-carlo tree search in continuous action spaces with value gradients
Chen et al. Solving multi-objective optimization problem using cuckoo search algorithm based on decomposition
Dong et al. Intelligent trainer for dyna-style model-based deep reinforcement learning
Goel et al. Improved multi-ant-colony algorithm for solving multi-objective vehicle routing problems
CN110850893A (en) Spacecraft task planning method based on maximum cost evaluation
CN114815801A (en) Adaptive environment path planning method based on strategy-value network and MCTS
Li et al. Development of a multi-objective artificial tree (MOAT) algorithm and its application in acoustic metamaterials
CN111190711B (en) BDD combined heuristic A search multi-robot task allocation method
CN112700099A (en) Resource scheduling planning method based on reinforcement learning and operation research
Koga et al. Speeding-up reinforcement learning through abstraction and transfer learning
CN115759199A (en) Multi-robot environment exploration method and system based on hierarchical graph neural network
Lee et al. A distributed dynamic framework to allocate collaborative tasks based on capability matching in heterogeneous multi-robot systems
Hu et al. Research on AGV Path Based on Optimal Planning
Wei et al. Long-term autonomous mission planning of AUV in large-scale complex marine environment
Ayub et al. Exhaustive study of essential constraint satisfaction problem techniques based on N-Queens problem
Xie et al. How autonomy oriented computing (AOC) tackles a computationally hard optimization problem
Zhang Enviroment Representations with Bisimulation Metrics for Hierarchical Reinforcement Learning
Bard et al. Necessary conditions for the linear three level programming problem
CN112348361B (en) Heuristic spacecraft task planning method based on state transition path reconstruction
CN116108613B (en) Method, system, equipment and medium for quickly constructing interpretable equipment combination

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