CN114240169A - Intelligent tactical decision method based on killer chain - Google Patents
Intelligent tactical decision method based on killer chain Download PDFInfo
- Publication number
- CN114240169A CN114240169A CN202111558502.2A CN202111558502A CN114240169A CN 114240169 A CN114240169 A CN 114240169A CN 202111558502 A CN202111558502 A CN 202111558502A CN 114240169 A CN114240169 A CN 114240169A
- Authority
- CN
- China
- Prior art keywords
- enemy
- combat
- analysis
- decision
- model
- 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 47
- 230000009471 action Effects 0.000 claims abstract description 103
- 238000004458 analytical method Methods 0.000 claims abstract description 64
- 238000012163 sequencing technique Methods 0.000 claims abstract description 4
- 230000005484 gravity Effects 0.000 claims description 50
- 238000011156 evaluation Methods 0.000 claims description 19
- 238000010586 diagram Methods 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 11
- 238000003012 network analysis Methods 0.000 claims description 8
- 230000000694 effects Effects 0.000 claims description 6
- 230000008447 perception Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 230000010485 coping Effects 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 230000007123 defense Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 230000010354 integration Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 125000004122 cyclic group Chemical class 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/29—Graphical models, e.g. Bayesian networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/451—Execution arrangements for user interfaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/466—Transaction processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06316—Sequencing of tasks or work
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an intelligent tactical decision method based on a killer chain, which relates to the technical field of air combat systems and comprises the steps of threat assessment and sequencing, enemy intention identification, our party task analysis, enemy combat center-of-gravity analysis, my combat center-of-gravity analysis, enemy combat action scheme prediction, generation of my combat action scheme and the like. Compared with the prior art, the method can solve the problems of more killer chain elements, large battlefield resource quantity and high system operation efficiency requirement, remarkably improve the auxiliary command decision-making capability of the command control system, and provide a timely and accurate auxiliary command decision-making scheme for a commander.
Description
Technical Field
The invention relates to the technical field of air combat systems, in particular to an intelligent tactical decision method based on a killer chain.
Background
In the context of joint combat, multi-domain collaboration has become one of the core capability elements in determining the success or failure of a war. Modern war has presented the characteristic of systematic confrontation, and through close cooperation and effective control among different forces and actions, an overall resultant force is formed, and the victory of battle is seized. The importance of system cooperation is emphasized by NIFC-CA (naval integrated air defense), global combined combat, mosaic and a plurality of combined combat theories newly proposed by the US army. Emerging technologies such as informatization and intellectualization promote new quality combat elements under system cooperation.
The killing chain is a series of cyclic elements forming attack and defense countermeasures in the battle, and the core of the air combat defeat lies in the rapid closed loop of the killing chain. The most representative killer chain model is the OODA theory proposed by Boider, and comprises 4 links of observation (observer) -judgment (Orient) -decision (Decide) -action (Act), and the model has become a methodology widely applied to air combat technical research at home and abroad at present. Because the air battle has the characteristics of high dynamic and strong game, the closed loop of the killing chain has a large amount of uncertainty, and an automatic and mechanical system is difficult to form. Under the backgrounds of combined combat, integrated air defense, multi-domain cooperation and the like, a killing chain in the air combat process is integrated with more resource cooperation, planning and optimization problems, and the plan of combat actions becomes a complex system engineering.
Of the 4 links of the OODA killing chain, the biggest impact on the closed-loop efficiency lies in the decision (origin) and decision (decision), both of which involve decision problems under conditions of complex information (covering external perception and internal duties). The intelligent algorithm has the advantages of processing mass information and learning of a complex system, so that the intelligent algorithm has the advantages of solving the judgment and decision problems.
At present, intelligent algorithm design is developed under the conditions of determined information and limited situation of intelligent air combat, and less information evaluation, prejudgment and combined action plan of the whole killing chain are involved. The information is determined by assuming that input information in a battlefield is accurate and only emphasizing the game of a decision-making link; the limited situation means that the complexity of air combat is limited to a relatively definite and small scale, and only the behavior decision of the engagement level is emphasized. The application background of the invention is combined combat and multi-domain cooperation, battlefield environment designs air, ground/sea surface combat resources, and the generation of the combat action plan needs to comprehensively consider the elements such as equipment capacity, formation situation, tactical procedures and the like. In summary, the problem of action plan generation under the background of combined combat is difficult to solve by intelligent decision in the field of air combat currently, and the following limitations mainly exist:
1) the factors of the killing chain are not completely considered, and most of intelligent decisions are used in decision loops of OODA (on-off object data acquisition), such as attack, defense and maneuver decisions, electronic countermeasure pattern decisions and the like;
2) the knowledge modeling of multi-domain (land, sea, air and sky) battlefield resources is insufficient, a large amount of domain knowledge is needed for support in the generation of a combat action plan, and the knowledge needs to be classified, managed and applied in the modeling process;
3) the operation efficiency of the algorithm is difficult to meet the operation requirement, the number of combat resources related to a multi-domain battlefield is large, and the operation amount of the algorithm is increased along with the number of the combat resources.
Disclosure of Invention
The intelligent tactical decision-making method based on the killer chain can solve the problems of more killer chain elements, large battlefield resource quantity and high system operation efficiency requirement, obviously improves the auxiliary command decision-making capability of a command control system, and provides a timely and accurate auxiliary command decision-making scheme for a commander.
In order to achieve the above purpose, the embodiment of the invention adopts the following technical scheme:
an intelligent tactical decision method based on a killer chain is characterized by comprising the following implementation steps:
step 1) threat assessment and ranking: deducing the threat degree of the enemy to the enemy according to the battlefield situation of the enemy and the my, carrying out threat priority ranking on the attacking targets of the enemy at the levels of speed, threat type, threat capability and target priority, analyzing the battlefield environment situation information by using a multi-body kinematics analysis method, clustering the enemy targets by using an improved K-means method, and finishing the ranking according to a threat degree evaluation index system;
step 2) enemy intention identification: on the basis of situation information and threat sequencing, comprehensively considering the space-time relationship and the field knowledge to explain the current situation, modeling the complex logical relationship among a plurality of random events into a directed acyclic graph and a group of probability distribution relationship descriptions, judging the battlefield deployment and action attempt of an enemy by using a Dynamic Bayesian Network (DBN) algorithm, and identifying the intention of the enemy, wherein: the factors analyzed by the algorithm include: target motion status, battlefield events, operational ordinances, tactical principles, and node status.
Step 3) analysis of tasks of the my party: analyzing the coping strategy of our party according to the command intention, the enemy intention and the enemy weapon efficiency of our party, applying an integrated definition model, and decomposing the mission task of our party in a graphical language, wherein the task comprises an adopted task plan and the number of weapons to be used;
step 4), analyzing the center of gravity of enemy combat: analyzing the hitting gravity center of the enemy to the enemy, namely a target of preferential hitting according to the enemy intention and the influence probability of the enemy to the enemy, combining a gravity center-key capability-key requirement-key weakness analysis mode of a hierarchical structure with a network analysis method ANP (artificial neural network) by using a multi-target network gravity center model, constructing a dependency relationship among all elements of a battlefield, namely a multi-target network diagram, and solving the fighting network gravity center based on a multi-entity Bayesian network;
step 5) carrying out operation gravity center analysis on one party: analyzing the hitting gravity center of the enemy of the party, namely a target of preferential hitting according to the task analysis and the mutual influence probability of the party, combining a gravity center-key capability-key requirement-key weakness analysis mode of a hierarchical structure with a network analysis method ANP (artificial neural network) by using a multi-target network gravity center model, constructing a dependency relationship among all elements of a battlefield, namely a multi-target network diagram, and solving the fighting network gravity center based on a multi-entity Bayesian network;
step 6) predicting the action scheme of the enemy battle: generating a feasible combat action scheme of the enemy according to the enemy intention and the combat barycenter analysis result and the enemy view angle, wherein: the method comprises the steps of abstracting the fighting dynamics into a series of state spaces, generating an enemy fighting action decision point and a decision line according to the results of enemy task analysis and fighting gravity center analysis, describing and modeling a fighting action scheme based on a Bayesian network, generating a feasible fighting action scheme solution space by using a planning algorithm, displaying the generated enemy fighting action scheme in a Gantt chart mode, wherein each row represents an action sequence of a fighting unit, and each column represents all executed actions in the current time slice;
step 7), generating a battle action scheme of the parties: developing a knowledge base model based on task requirements and situation prediction; modeling a combat ontology, wherein the models comprise a combat task model, a combat unit model, a threat event model, a combat resource model and a constraint condition model; abstracting the fighting dynamics into a series state space, generating a fighting action decision point and a decision line of a party according to the results of task analysis and fighting gravity center analysis of the party, describing and modeling a fighting action scheme based on a Bayesian network, and generating a feasible fighting action scheme solution space by using a planning algorithm; the generated action scheme of the I-party battle is displayed in the form of a Gantt chart, each row represents the action sequence of the battle unit, and each column represents all executed actions in the current time slice;
step 8), carrying out deduction and evaluation: based on action matrixes of the two parties of the enemy and the my, deduction and evaluation of a feasible combat action scheme of the party is completed based on a combat effect, and the deduction and evaluation criterion of the combat action scheme comprises the following steps: maneuver scale, kill level, loss forecast, resource consumption, task flexibility, and risk level.
Preferably, in the enemy intention identifying step: and (3) constructing a dynamic Bayesian network intention recognition model by using a statistic-based uncertain reasoning method, and realizing the recognition analysis of the enemy intention.
Preferably, in the battle center-of-gravity analysis step: a multi-target network gravity center model is provided by applying a hierarchical structure analysis mode and a network analysis method, the multi-target network gravity center model respectively carries out probabilistic reasoning solution on the enemy combat gravity center and the my combat gravity center, and the battlefield key combat gravity center analysis is completed, wherein: the multi-target network gravity center model comprises three views: the method comprises the following steps of (1) obtaining a multi-target network situation view, a COG-CC-CR-CV view and a COG network view; the multi-target network gravity center model comprises five steps: target decomposition, dependency relationship analysis, COG network construction, importance calculation and evolution analysis.
Preferably, in the my party combat action plan generating step:
through constructing a basic action model, a killing chain model, a decision point model and a battle line model, an optimal battle action scheme of our party is planned and optimized, and an intelligent tactic decision is completed, wherein: basic combat actions are the basis for constructing a combat action scheme model of our part, all available basic combat actions of our part are input by a task analysis model of our part, required combat resources and time are determined by expert knowledge, and the basic actions can define the preorder actions of the basic actions, define the effect of action influence and the probability of influence, define the area for action execution and define the required combat resources according to a killing chain when the basic actions are initialized; the killing chain model makes a corresponding fighting killing chain for each fighting unit, and provides rule constraint for the establishment of a basic feasible action database for the subsequent time segment division.
Preferably, the operation deduction evaluation step comprises a Wargaming function, and the Wargaming function can carry out operation deduction evaluation on the operation action scheme generated by intelligent decision, simulate the operation action scheme of the local party and the operation action scheme of the enemy party, and determine the advantages and disadvantages of the operation action scheme of the local party and places needing attention in future operations.
Preferably, based on the decision service requirement of the multi-domain battlefield environment, the decision service requirement is processed in parallel according to 3 threads to realize that the decision requirements of different levels synchronously meet the task requirement, wherein: calculating situation perception classes according to 50ms levels in real time, wherein the situation perception comprises intention identification and task analysis; calculating situation prediction according to 2000 ms-level low delay, wherein the situation prediction comprises enemy combat center of gravity generation, enemy combat center of gravity generation and enemy combat action scheme prediction; and calculating and pushing auxiliary decision classes according to the limited delay of 10s grade, wherein the auxiliary decision comprises the generation of a combat action scheme and the deduction of a combat.
The intelligent tactical decision method based on the killer chain has the following beneficial effects:
1) the invention completely considers the elements of the killing chain in the combined operation and is more suitable for a multi-domain cooperative application scene. Based on the multi-domain combined combat background, situation prediction, evaluation and plan decision making are completed according to the flow by an 8-step hierarchical decision method, and are subdivided into enemy intention identification, my party task analysis, enemy COG generation, my party COG generation, enemy COA prediction, my party COA generation and combat deduction, and judgment (origin) and decision (Decide) elements in a killing chain are considered more completely.
2) The invention decouples the intelligent algorithm from the domain knowledge, and is more suitable for engineering design and integrated application. Based on battlefield modeling and decision layering, the relative determination of intelligent decision problem domains of different steps/modules is realized, the decoupling of a multi-entity Bayesian network and a battle resource behavior knowledge base is realized, a working interface of an algorithm tool and knowledge modeling is formed, and the engineering application of an intelligent decision method is facilitated.
3) The method has higher operation efficiency, and can generate/push prediction, evaluation and decision results in parallel according to different service requirements. Based on the decision service requirements of the multi-domain battlefield environment, parallel processing is carried out according to 3 threads, wherein situation perception classes (intention identification and task analysis) are operated in real time (50ms level), situation prediction classes (enemy COG generation, our COG generation and enemy COA prediction) are operated in low delay (2000ms level), auxiliary decision classes (COA generation and combat deduction) are operated in limited delay (10s level) and pushed, and the decision requirements of different levels can be synchronously met with the task requirements.
Drawings
FIG. 1 is a schematic overall workflow of the present invention;
FIG. 2 is a schematic diagram of a three-thread parallel integration implementation of the present invention;
FIG. 3 is a schematic diagram of a Bayesian network with enemy intent identification of the present invention;
FIG. 4 is a schematic diagram of a COG model processing flow according to the present invention;
FIG. 5 is a schematic illustration of the COA prediction process of the enemy of the present invention;
FIG. 6 is a schematic diagram of the IDEF model for my party task analysis in accordance with the present invention;
FIG. 7 is a schematic diagram of the COA generation process of my party of the present invention;
FIG. 8 is a schematic flow chart of the Wargaming function of the present invention.
Detailed Description
The invention will be further explained with reference to the accompanying drawings in which:
as shown in fig. 1 and 8, the invention provides an intelligent generation method of an action plan for a multi-domain combined combat environment, and mainly solves the problems of multiple elements of a killer chain, large quantity of battlefield resources and high requirement on system operation efficiency. In order to improve the auxiliary command decision-making capability of the command control system, an intelligent decision-making system is utilized to provide a timely and accurate auxiliary command decision-making scheme for a commander. The method consists of a series of operation intentions, operation gravity centers and action prediction subfunction modules, wherein the analysis result of the subfunction is the analysis result of the battlefield situation and the input parameter of the downstream subfunction analysis, and the final output of the intelligent tactical decision method based on the killing chain is the COA of the operation action scheme of our party.
The intelligent tactical decision method based on the killer chain is implemented by 8 steps in total:
step 1: threat assessment and ranking: according to the battlefield situation of the enemy and the my, the threat degree of the enemy to the enemy is deduced, the threat priority ranking is carried out on the attacking targets of the enemy at multiple levels such as speed, threat type, threat capability and target priority, the battlefield environment situation information is analyzed by using a multi-body kinematics analysis method, the grouping of the enemy targets is realized by using an improved K-means method, and the ranking is completed according to the threat degree evaluation index system.
Step 2: enemy intention identification: on the basis of situation information and threat sequencing, the current situation is explained by comprehensively considering the spatiotemporal relationship and the domain knowledge. Modeling a complex logic relationship among a plurality of random events into a directed acyclic graph and a group of probability distribution relationship descriptions, judging battlefield deployment and action attempts of an enemy by using a Dynamic Bayesian Network (DBN) algorithm, and identifying the intention of the enemy. The factors analyzed by the algorithm include: target motion status, battlefield events, operational ordinances, tactical principles, and node status.
And step 3: and (3) analyzing tasks of the my party: according to the command intention, the enemy intention and the enemy weapon efficiency of the my party, the coping strategy of the my party is analyzed, and the integrated definition (IDEF model) is applied to decompose the fighting mission task of the my party by using a graphical language, wherein the mission comprises an adopted mission plan and the number of weapons to be used.
And 4, step 4: analysis of enemy COG: according to the intention of enemies and the influence probability among the enemies, the hitting center of the enemies to our party, namely the target of preferential hitting is analyzed, a multi-target network center-of-gravity model is applied, a center-of-gravity-key-requirement-key weakness analysis mode of a hierarchical structure is combined with a network analysis method ANP, the dependency relationship among all elements of a battlefield is constructed, namely a multi-target network diagram, and the fighting network center of gravity is solved based on a multi-entity Bayesian network MEBN.
And 5: my party COG analysis: and (4) analyzing the hitting gravity center of the enemy of the party, namely the target of the prior hitting according to the task analysis of the party and the influence probability of the party and the enemy, and performing the same algorithm as the step 4.
Step 6: enemy COA prediction: and generating feasible COA of the enemy according to the enemy intention and the COG analysis result and in an enemy view. The algorithm is the same as step 7.
And 7: the COA of my party is generated: developing a knowledge base model based on task requirements and situation prediction; and modeling a combat ontology, wherein the models comprise a combat task model, a combat unit model, a threat event model, a combat resource model and a constraint condition model. The fighting dynamic state is abstracted into a series of state spaces, a fighting action decision point and a decision line of a party are generated according to the task analysis and COG analysis results of the party, the COA is described and modeled based on a Bayesian network, and a feasible COA solution space is generated by applying a planning algorithm. The generated my COA is shown in the form of a Gantt chart, each row representing the sequence of actions for the engagement unit, each column representing all actions performed during the current time segment, and the Gantt chart example is shown in the following table.
And 8: and (3) performing deduction and evaluation: and based on action matrixes of the two sides of the enemy and the my, deduction evaluation of the feasible COA of the party is completed based on the fighting effect. The COA derived assessment criteria include: maneuver scale, kill level, loss forecast, resource consumption, task flexibility, and risk level.
In this embodiment, a functional logic architecture of a complete intelligent tactical decision method based on a killer chain is shown in fig. 1, a program implementation of the method adopts three-thread parallel operation, and a technical integration implementation scheme is shown in fig. 2. The threat assessment and sorting function achieves grouping of enemy targets and sorting according to the threat degree by analyzing the situation information of the battlefield environment. The enemy intention recognition function constructs a dynamic Bayesian network intention recognition model by using a statistical-based uncertain reasoning method, so as to realize enemy intention recognition analysis, wherein the enemy intention recognition Bayesian network is shown in FIG. 3.
In the embodiment, in order to identify the battle center of gravity COG in a complex battlefield environment and simultaneously protect the key weakness CV of our party, the Analytic function of the enemy/our party COG applies a hierarchical structure analysis mode and a Network analysis method Analytic Network Progress, the ANP provides a multi-target Network center of gravity model, probabilistic reasoning solution is respectively carried out on the enemy COG and the our party COG, and the battlefield key battle center of gravity analysis is completed, wherein a flow diagram of the GOG model is shown in fig. 4.
Specifically, the model mainly comprises three views: the method comprises the following steps of (1) obtaining a multi-target network situation view, a COG-CC-CR-CV view and a COG network view; the method comprises five steps of target decomposition, dependency relationship analysis, COG network construction, importance calculation and evolution analysis. In order to predict possible actions of an enemy under the current situation, an enemy COA prediction function predicts an enemy combat action scheme COA by constructing an enemy decision point and a combat line model and utilizing an enemy COG and an enemy map analysis result, and a flow block diagram of an enemy COA prediction sub-function is shown in FIG. 5.
In this embodiment, the analysis of the tasks of our party uses an Integration DEFinition, IDEF model, and uses a graphical language, a symbolic abstraction and a modularization principle to decompose the fighting mission of our party into the fighting tasks that can be executed. The IDEF model map for my task analysis is shown in fig. 6.
In this embodiment, the COA generation function of our party plans and optimizes the optimal combat action scheme of our party by constructing a basic action model, a killer chain model, a decision point model and a combat line model, thereby completing an intelligent tactical decision. Basic combat actions are the basis for building the COA model of our party, all available basic combat actions of our party are input by the mission analysis model of our party, and the required combat resources and time are determined by expert knowledge. The basic action defines the prior action according to the killing chain, defines the effect and the influence probability of the action influence, defines the execution area of the action and defines the required combat resources. The killing chain model establishes a corresponding operational killing chain OODA for each operational unit, and provides rule constraint for the establishment of a basic feasible action database for the subsequent time segment division. The flow diagram of the my COA generation sub-function is shown in fig. 7.
It should be noted that the wargamg function can develop a tactical deduction evaluation on the tactical action scheme generated by the intelligent decision. The COA of our party and the COA of the enemy are simulated to fight, and the advantages and the disadvantages of the COA of our party and places needing attention in future battles are determined. The flow chart of the Wargaming function is shown in fig. 8.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (6)
1. An intelligent tactical decision method based on a killer chain is characterized by comprising the following implementation steps:
step 1) threat assessment and ranking: deducing the threat degree of the enemy to the enemy according to the battlefield situation of the enemy and the my, carrying out threat priority ranking on the attacking targets of the enemy at the levels of speed, threat type, threat capability and target priority, analyzing the battlefield environment situation information by using a multi-body kinematics analysis method, clustering the enemy targets by using an improved K-means method, and finishing the ranking according to a threat degree evaluation index system;
step 2) enemy intention identification: on the basis of situation information and threat sequencing, comprehensively considering the space-time relationship and the field knowledge to explain the current situation, modeling the complex logical relationship among a plurality of random events into a directed acyclic graph and a group of probability distribution relationship descriptions, judging the battlefield deployment and action attempt of an enemy by using a Dynamic Bayesian Network (DBN) algorithm, and identifying the intention of the enemy, wherein: the factors analyzed by the algorithm include: target motion state, battlefield events, combat ordinances, tactical principles, and node state;
step 3) analysis of tasks of the my party: analyzing the coping strategy of our party according to the command intention, the enemy intention and the enemy weapon efficiency of our party, and decomposing the fighting mission task of our party by using a graphical language by using an Integrated Definition (IDEF) model, wherein the task comprises an adopted task plan and the number of weapons to be used;
step 4), analyzing the center of gravity (COG) of enemy combat: analyzing the hitting gravity center of the enemy to the enemy, namely a target of preferential hitting according to the enemy intention and the influence probability of the enemy to the enemy, combining a gravity center-key capability-key requirement-key weakness analysis mode of a hierarchical structure with a network analysis method ANP (artificial neural network) by using a multi-target network gravity center model, constructing a dependency relationship among elements of a battlefield, namely a multi-target network diagram, and solving the fighting network gravity center based on a multi-entity Bayesian network (MEBN);
step 5) analysis of the center of gravity (COG) of the battle of the party: analyzing the hitting gravity center of the enemy of the party, namely a target of preferential hitting according to the task analysis and the mutual influence probability of the party, combining a gravity center-key capability-key requirement-key weakness analysis mode of a hierarchical structure with a network analysis method ANP (artificial neural network) by using a multi-target network gravity center model, constructing a dependency relationship among elements of a battlefield, namely a multi-target network diagram, and solving the fighting network gravity center based on a multi-entity Bayesian network (MEBN);
step 6) enemy combat action plan (COA) prediction: generating a feasible combat action plan (COA) of the enemy from the perspective of the enemy according to the analysis result of the enemy intention and the center of gravity (COG) of the combat, wherein: abstracting the battle dynamics into a series state space, generating enemy battle action decision points and decision lines according to the results of enemy task analysis and battle center of gravity (COG) analysis, describing and modeling a battle action scheme (COA) based on a Bayesian network, and generating a feasible battle action scheme (COA) solution space by using a planning algorithm, wherein the generated enemy battle action scheme (COA) is displayed in a Gantt diagram form, each row represents an action sequence of the battle unit, and each column represents all executed actions in the current time slice;
step 7) generating a fighting action scheme (COA) of the party: developing a knowledge base model based on task requirements and situation prediction; modeling a combat ontology, wherein the models comprise a combat task model, a combat unit model, a threat event model, a combat resource model and a constraint condition model; abstracting the fighting dynamics into a series state space, generating a fighting action decision point and a decision line of a party according to the task analysis and fighting gravity Center (COG) analysis results of the party, describing and modeling a fighting action scheme (COA) based on a Bayesian network, and generating a feasible fighting action scheme (COA) solution space by using a planning algorithm; the generated our party combat action scheme (COA) is shown in the form of a Gantt chart, each row represents the action sequence of the combat unit, and each column represents all executed actions in the current time slice;
step 8), carrying out deduction and evaluation: based on action matrixes of two parties of the enemy and the my, deduction evaluation of a feasible combat action scheme (COA) of one party is completed based on combat effects, and the deduction evaluation criterion of the combat action scheme (COA) comprises the following steps: maneuver scale, kill level, loss forecast, resource consumption, task flexibility, and risk level.
2. The intelligent tactical decision-making method based on killer chain as claimed in claim 1 wherein, in said enemy intention identifying step: and (3) constructing a dynamic Bayesian network intention recognition model by using a statistic-based uncertain reasoning method, and realizing the recognition analysis of the enemy intention.
3. The intelligent tactical decision method based on killer chain as claimed in claim 1, wherein in the battle center of gravity (COG) analysis step: a multi-target Network gravity center model is provided by applying a hierarchical structure analysis mode and a Network Analysis (ANP), the multi-target Network gravity center model respectively carries out probabilistic reasoning solution on the enemy combat gravity Center (COG) and the my combat gravity Center (COG), and the critical combat gravity center analysis in a battlefield is completed, wherein:
the multi-target network gravity center model comprises three views: the method comprises the following steps of (1) obtaining a multi-target network situation view, a COG-CC-CR-CV view and a COG network view; the multi-target network gravity center model comprises five steps: target decomposition, dependency relationship analysis, COG network construction, importance calculation and evolution analysis.
4. The intelligent tactical decision method based on killer chain as claimed in claim 1, wherein in said generation step of my party operational action scheme (COA):
through constructing a basic action model, a killing chain model, a decision point model and a battle line model, an optimal battle action scheme of our party is planned and optimized, and an intelligent tactic decision is completed, wherein: basic combat actions are the basis of constructing a combat action scheme (COA) model of our part, all available basic combat actions of our part are input by a mission analysis model of our part, required combat resources and time are determined by expert knowledge, and the basic combat actions define the preorder actions of the basic combat actions, define the effect and the probability of the influence of the actions, define the execution area of the actions and define the required combat resources according to a killer chain when being initialized; the killing chain model establishes a corresponding combat killing chain (OODA) for each combat unit, and provides rule constraint for the establishment of a basic feasible action database for the subsequent time segment division.
5. The intelligent tactical decision-making method based on the killer chain as claimed in claim 1, wherein the tactical deduction evaluation step comprises a Wargaming function, the Wargaming function can carry out tactical deduction evaluation on the tactical action scheme generated by the intelligent decision, the tactical action scheme (COA) of our party and the hostile party tactical action scheme (COA) are simulated to fight, and the advantages and disadvantages of the tactical action scheme (COA) of our party and places needing attention in the future battle are determined.
6. The intelligent tactical decision-making method based on the killer chain as claimed in claim 1, wherein based on the decision-making service requirement of the multi-domain battlefield environment, the decision-making service requirement is processed in parallel according to 3 threads to realize that the decision-making requirements of different levels synchronously meet the task requirement, wherein:
calculating situation perception classes according to 50ms levels in real time, wherein the situation perception comprises intention identification and task analysis;
calculating situation prediction classes according to 2000 ms-level low delay, wherein the situation prediction comprises enemy combat center of gravity (COG) generation, my combat center of gravity (COG) generation and enemy combat action scheme (COA) prediction;
and calculating and pushing auxiliary decision classes according to the limited delay of 10s grade, wherein the auxiliary decision comprises the generation of a combat action scheme (COA) and the deduction of the combat.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111558502.2A CN114240169A (en) | 2021-12-17 | 2021-12-17 | Intelligent tactical decision method based on killer chain |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111558502.2A CN114240169A (en) | 2021-12-17 | 2021-12-17 | Intelligent tactical decision method based on killer chain |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114240169A true CN114240169A (en) | 2022-03-25 |
Family
ID=80758789
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111558502.2A Pending CN114240169A (en) | 2021-12-17 | 2021-12-17 | Intelligent tactical decision method based on killer chain |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114240169A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115952428A (en) * | 2022-12-31 | 2023-04-11 | 中国电子科技集团公司信息科学研究院 | GRU-based group task identification method |
| CN116680542A (en) * | 2023-06-27 | 2023-09-01 | 北京五木恒润科技有限公司 | Method and system for generating trunk branch situation and bypass branch situation |
-
2021
- 2021-12-17 CN CN202111558502.2A patent/CN114240169A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115952428A (en) * | 2022-12-31 | 2023-04-11 | 中国电子科技集团公司信息科学研究院 | GRU-based group task identification method |
| CN115952428B (en) * | 2022-12-31 | 2023-11-14 | 中国电子科技集团公司信息科学研究院 | Group task identification method based on GRU |
| CN116680542A (en) * | 2023-06-27 | 2023-09-01 | 北京五木恒润科技有限公司 | Method and system for generating trunk branch situation and bypass branch situation |
| CN116680542B (en) * | 2023-06-27 | 2024-01-09 | 北京五木恒润科技有限公司 | Method and system for generating trunk branch situation and bypass branch situation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112329348B (en) | Intelligent decision-making method for military countermeasure game under incomplete information condition | |
| CN114240169A (en) | Intelligent tactical decision method based on killer chain | |
| CN113392521B (en) | Method and system for constructing resource marshalling model for air-sea joint combat mission | |
| CN112742028B (en) | Formation decision method, system, medium and equipment for countermeasure game | |
| Biltgen et al. | A Methodology for Capability-Focused Technology Evaluation of Systems of Systems | |
| CN113298260A (en) | Confrontation simulation deduction method based on deep reinforcement learning | |
| CN110109653A (en) | A kind of land battle war game intelligent engine and its operation method | |
| Erdal et al. | Evaluation of Anti-Tank Guided Missiles: An integrated Fuzzy Entropy and Fuzzy CoCoSo multi criteria methodology using technical and simulation data | |
| Brynielsson | Using AI and games for decision support in command and control | |
| CN112287521A (en) | Decision-making platform of intelligent combat equipment | |
| Azak et al. | A new approach for Threat Evaluation and Weapon Assignment problem, hybrid learning with multi-agent coordination | |
| CN117272835A (en) | Digital twin soldier chess deduction meteorological data simulation and intelligent decision-making method and system | |
| EP1666842A1 (en) | A system and a method for mission planning | |
| Yuksek et al. | Intelligent Wargaming Approach to Increase Course of Action Effectiveness in Military Operations | |
| Sisti | Dynamic situation assessment and prediction (DSAP) | |
| Zhang et al. | Intelligent battlefield situation comprehension method based on deep learning in wargame | |
| Najgebauer et al. | The automation of combat decision processes in the simulation based operational training support system | |
| Kebir et al. | A multi-agent model for countering terrorism | |
| CN114529205B (en) | Mosaic warfare system capacity demand satisfaction evaluation method and related equipment | |
| Johnson et al. | Mapping artificial intelligence to the naval tactical kill chain | |
| Shi et al. | The UAV swarm mission planning based on adaptive genetic algorithm | |
| Plamondon | A frigate survival approach based on real-time multiagent planning | |
| Jin et al. | Research on Hybrid Intelligence Wargame Method | |
| Butler et al. | Predicting the movements of robot teams using generative models | |
| McQuay et al. | Distributed collaborative decision support environments for predictive awareness |
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 |