CN114282833A - Rules-based hierarchical task planning method for air-sea combined combat action - Google Patents
Rules-based hierarchical task planning method for air-sea combined combat action Download PDFInfo
- Publication number
- CN114282833A CN114282833A CN202111630161.5A CN202111630161A CN114282833A CN 114282833 A CN114282833 A CN 114282833A CN 202111630161 A CN202111630161 A CN 202111630161A CN 114282833 A CN114282833 A CN 114282833A
- Authority
- CN
- China
- Prior art keywords
- planning
- task
- mission
- combat
- battle
- 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
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a rule-based hierarchical task planning method for air-sea combined combat actions, and belongs to the technical field of task planning. The invention designs a rule-based hierarchical task planning method based on the island shelter attacking and defending fighting scenario as the background, can intelligently analyze and mine the acquired battlefield data, and excavates deep information and characteristics from the data, thereby providing support for decision making.
Description
Technical Field
The invention belongs to the technical field of task planning, and particularly relates to a rule-based hierarchical task planning method for air-sea combined combat actions.
Background
The task planning is an important function of the C4ISR system, and essentially adopts an operation planning and optimizing method and idea to comprehensively plan resources and complete tasks. The mission planning technology can utilize new technologies such as knowledge/rule base, big data, cloud computing and artificial intelligence to carry out battlefield situation intelligent analysis and prediction, acquire enemy information, own situation information and battlefield environment information including army deployment, operation situation, activity rules and other information, intelligently analyze and mine the acquired battlefield data, mine deep information and characteristics from the acquired battlefield data, and provide support for decision making.
With the diversification of modern war forms and the complication of battlefield situations, a large amount of war deduction simulation and attack and defense countermeasures need to be carried out at ordinary times and in war, the war process in the deduction simulation is analyzed, military theories in the aspects of command decision, war planning and the like are further researched, the war knowledge is extracted from the simulation deduction data and is applied to the war, and therefore the purpose of learning the war from the war is achieved.
Disclosure of Invention
Technical problem to be solved
The technical problem to be solved by the invention is as follows: how to design a hierarchical task planning method for sea-air combined combat actions is to analyze and mine acquired battlefield data, and mine deep information and characteristics from the battlefield data so as to provide support for decision making.
(II) technical scheme
In order to solve the technical problems, the invention provides a rule-based sea-air combined combat action layered task planning method, which decomposes a sea-air battle process for island reef attack and defense planning into a high-level battle process and a low-level tactic-level battle process according to an OODA (object oriented data acquisition) ring model;
the high-level battle-level combat process is represented by an OODA (object oriented data acquisition) ring with a long period, and means that battle field situations are judged by collecting battle field information under the constraint of a specific mission task and combat capability, so that the planning of the battle tasks and the issuing of combat instructions are realized;
the low-level tactical combat process is represented by using an OODA ring with a short period, and means that on the basis of evaluating the combat capability and threat situation according to an instruction issued by a superior level, the local battlefield information is collected and the local battlefield situation is judged, so that the mission planning and tactical decision of basic combat marshalling or single-package basic action are realized; the low-level tactical combat process runs in an OODA ring of the high-level tactical combat process, the OODA ring of each low-level tactical combat process represents a subtask, the execution result of all the subtasks changes the battlefield situation to influence the high-level tactical combat process, and judgment and decision are taken as the content of command and decision activities and are carried out in a rule reasoning mode.
Preferably, according to the hierarchical characteristics of the sea-air combat process, a rule-based hierarchical mission planning model structure is constructed to describe the sea-air combat process conceived for the island reef attack and defense, the model structure comprises two levels of a battle mission planning model and a tactic mission planning model, the two levels respectively correspond to a high-level battle level combat process and a low-level tactic level combat process, and the model structure provides a situation perception interface and an instruction transmission interface which are respectively used for acquiring battlefield situation data and issuing mission instructions.
Preferably, in the method, the campaign mission planning layer is designed as follows: reflecting the process of carrying out operation planning according to the operation purpose and mission task of a commander under the global situation, and the realized specific planning flow comprises the following stages: mission task decomposition, formation resource distribution and force generation, finite-state machine-based battle layer mission planning and formation mission instruction generation;
the tactical mission planning layer is designed as follows: reflecting the collaborative planning of the battle formation, solidifying the relevant knowledge and collaborative method of tactical planning in the bottom planning model of the tactical mission planning layer, and forming the corresponding planning result on the basis of the battlefield situation and the upper-layer instruction; the realized specific planning process comprises the following stages: and analyzing the formation task instruction, planning the tactical layer task based on hard coding, and generating a single-package instruction.
Preferably, in the method, the situational awareness interface is designed to: the layered battle planning intelligent body provides battlefield situation data and is packaged with an interactive interface of the environment, and detection data, war newspaper data, battle damage data and equipment state data can be extracted from a battle simulation environment; the instruction transmission interface is designed as an ICE instruction transmission interface, the ICE instruction transmission interface is an interactive interface packaged for transmitting the formation task instruction and the single-loading action instruction, and an ICE protocol is applied to support instruction transmission between different levels and instruction transmission between a task planning intelligent agent and a simulation deduction environment.
Preferably, in the method, the mission task decomposition stage in the planning flow of the campaign mission planning layer is designed as: according to mission task targets and combat resources, tasks are decomposed into regional air defense tasks, early warning reconnaissance tasks, maritime patrol tasks, interception and attack tasks, high-price aerial asset protection, withdrawal defense and return voyage supply;
the regional air defense task is a battle-level task for assigning equipment such as fighters, ships and ground air defense weapons to form a fire power network to defend a preset key airspace, and aims to prevent enemies from performing air defense; the early warning reconnaissance task is a task of assigning a fighter plane and an early warning plane to perform reconnaissance detection in a specified airline or area, and aims to provide detection information and data for own parties; the maritime patrol task is a task of dispatching a guard ship and carrying out battle patrol on equipment in a designated navigation line or area by a submarine, and aims to provide detection capability support and firepower support and prevent an enemy from defending from the sea; the interception and striking task is a task of sending a combat aircraft, a ship and ground air defense facilities to intercept and strike local equipment for penetration defense; the high-value aerial asset protection task is a task of assigning a fighter to protect a high-value target, namely an early warning aircraft; withdrawal defense is a task that under the disadvantage, the unit of our party withdraws a security area to carry out defense combat; the return voyage supply refers to a task of returning to a base for maintenance and supply under the condition that the ammunition amount and the oil amount are insufficient or damaged.
Preferably, in the method, the formation resource allocation and power generation stage in the planning flow of the campaign mission planning layer is designed as follows:
and (3) queuing resource allocation: dividing the combat resources into different combinations for completing adaptive tasks; adopting a manual marshalling mode to divide combat resources which want to be centered into 12 marshalling;
force generation: constructing a battle marshalling into an Agent object, wherein each Agent can execute different tasks; constructing different Agent objects, and obviously constructing 6 types of Agent objects according to combat scenarios; each type of Agent consists of static attribute information and a functional module;
preferably, in the method, the stage of the finite-state-machine-based campaign-level mission planning in the planning flow of the campaign mission planning layer is designed as:
planning the operation: for a dynamic condition judgment, behavior planning and adjustment process, a finite state machine FSM is adopted to carry out regular organization and formation task instruction generation;
the finite-state-machine-based battle-level mission planning is developed for all Agent objects, and at the same time, according to battlefield situation conditions, mission decisions are performed on all Agent objects in parallel, and finally, mission decision results of all agents are combined to form a joint mission sequence as output of a battle-level mission planning layer;
according to the characteristics of the mission plan of the campaign layer, the mission plan of the campaign layer is represented by a simplified finite-state machine as follows:
Transitions＝(source,trigger,dest,after)
the source represents a task set which is currently executed, the task set is called as a state (state) in a state machine, trigger represents a trigger condition, the trigger is a task change judgment condition formed according to a situation, dest represents a next state, namely a task which needs to be executed after the trigger condition is met, and after is a series of operations which need to be executed for executing the task; transitions store the current state, change condition, and next state, which is another representation of the production rule IF-then structure.
Preferably, in the method, the formation task instruction generation phase in the planning flow of the campaign task planning layer is designed as:
after the battle-level tasks of the formation at the dynamic moment are obtained, the instruction set of the battle-level tasks of the formation is expressed as the following form:
CampaignSeqList＝[CampaignSeq1,CampaignSeq2,…,CampaignSeqN]
the specific task instruction CampiagnSeqi comprises the contents of task execution formation, task name, task type, task execution area and task target;
the output result, i.e., the set of tactical mission instructions, may then be packaged into a dictionary format for delivery to the tactical mission planning layer via the ICE protocol.
Preferably, in the method, a formation task instruction parsing stage in a planning flow of the tactical mission planning layer is designed as follows: and through data decoding and data extraction and analysis of a battle task instruction set, the formation tasks are extracted one by one and input into a tactic task planning layer.
Preferably, in the method, the tactical mission planning phase based on hard coding in the planning flow of the tactical mission planning layer is designed as:
the tactical mission planning layer is divided into 6 channels: planning regional air defense tasks, planning early warning and reconnaissance tasks, planning high-value aerial asset protection tasks, planning maritime patrol tasks, planning interception and attack tasks, planning formation return-voyage tasks, and synchronously planning tactical tasks of each formation;
after the task starts, battlefield situation acquisition, air combat decision and entity action output flow are carried out in sequence, situation evaluation analysis is firstly carried out before the air combat decision, the battlefield situation is evaluated and the air combat decision is carried out by using an IF-THEN rule, and the rules used in the regional air defense task are as follows:
(1) if the fighter team of the fighter plane is not in the air defense area or the airplane of the enemy is not detected, the fighter team of the fighter plane is maneuvered into the air defense area and patrolling and detecting are carried out;
(2) if the fighter team of the party is in the air defense area and the enemy plane is detected, but the enemy plane is not found to enter the early warning area of the party, the fighter team of the party executes the tracking and monitoring task but cannot leave the early warning area;
(3) if I put the fighter team in the air defense area, scout the enemy plane and find that the enemy plane enters the early warning area of the party, then judge that the enemy plane has an attempt to attack the I, and I fighter team executes the air-air combat mission and intercepts and attacks the attacking enemy plane;
the output result of the tactical mission planning based on hard coding is a formation tactical mission, which comprises formation area search, formation tracking enemy plane, formation take-off, formation maneuvering and formation air-to-air combat, which are respectively packaged as independent functions.
(III) advantageous effects
The invention designs a rule-based hierarchical task planning method based on the island shelter attacking and defending fighting scenario as the background, can intelligently analyze and mine the acquired battlefield data, and excavates deep information and characteristics from the data, thereby providing support for decision making.
Drawings
FIG. 1 is a schematic diagram of a rule-based hierarchical mission planning model according to the present invention;
FIG. 2 is a schematic diagram of a decision mechanism of the campaign level task planning layer of the present invention;
FIG. 3 is a schematic diagram of a finite state machine for the air defense behavior of the fighter team area depicted in a state transition diagram of the invention;
FIG. 4is a schematic diagram of the hard-coded tactical layer mission planning of the present invention;
fig. 5 is a flowchart of the regional air defense task of the present invention.
Detailed Description
In order to make the objects, contents, and advantages of the present invention clearer, the following detailed description of the embodiments of the present invention will be made in conjunction with the accompanying drawings and examples.
The invention designs a rule-based layered task planning method based on the island attack and defense fighting scenario, wherein a battle task planning layer mainly realizes the battle task planning of various formation and the formation task instruction generation; the tactical mission planning layer mainly realizes analysis of formation mission instructions, tactical mission planning of members (single-unit) in the formation and generation of action instructions which can be executed by the simulation deduction platform. The rule-based tactical mission planning layer is designed by using a finite-state machine method, and the rule-based tactical mission planning layer is designed by using a hard coding method. The specific invention content comprises:
(1) designing a hierarchical task planning model structure based on rules;
(2) designing a battle-level mission planning method based on a finite-state machine;
(3) the tactical task planning method based on hard coding is designed.
3 technical implementation approach
3.1 rule-based hierarchical mission planning model architecture design
According to the OODA ring model, the sea-air combat process for island reef attack and defense planning is decomposed into a high-level battle-level combat process and a low-level tactic-level combat process;
the high-level battle-level combat process is represented by an OODA (object oriented data acquisition) ring with a long period, which means that battle field situations are judged by collecting battle field information under the constraint of a specific mission task and combat capability, so that the planning of the battle task and the issuing of combat instructions are realized;
the low-level tactical combat process is represented by an OODA ring with a short period, which means that on the basis of evaluating the combat capability and threat situation according to an instruction issued by a superior level, the local battlefield information is collected and the local battlefield situation is judged, so that the mission planning and tactical decision of basic combat marshalling or single-package basic action are realized; the low-level tactical combat process runs in an OODA ring of the high-level tactical combat process, the OODA ring of each low-level tactical combat process represents a subtask, the execution results of all the subtasks change the battlefield situation to influence the high-level combat process, and judgment and decision are taken as the core of decision-making activity command and are carried out in a rule reasoning mode.
According to the layering characteristics of the air-sea combat process, a rule-based layered mission planning model structure is constructed, the model structure comprises a battle mission planning model and a tactic mission planning model, and the model structure provides a situation awareness interface and a command transmission interface which are respectively used for acquiring battlefield situation data and issuing mission commands. FIG. 1 is a schematic diagram of a hierarchical mission planning model structure.
Battle mission planning layer: the battle layer task planning reflects the process of battle planning according to the battle purpose and mission task by commanders under the global situation, and the specific planning flow comprises mission task decomposition, formation resource distribution and strength generation, battle layer task planning based on a finite state machine and formation task instruction generation;
tactical mission planning layer: the tactical mission planning reflects the collaborative planning of the battle formation, the relevant knowledge and the collaborative method of the tactical planning are solidified in a bottom layer planning model, and a corresponding planning result is formed on the basis of the battlefield situation and the upper layer instruction; the specific planning process comprises the following steps: analyzing the formation task instruction, planning the tactical layer task based on hard coding, and generating a single-mounted instruction;
the model structure comprises two types of interfaces as follows: (1) situation awareness interface: the situation awareness interface is an interactive interface packaged with the environment for providing battlefield situation data for the layered battle planning agent. Detection data, war newspaper data, war damage data, equipment state data and the like can be extracted from the combat simulation environment; (2) ICE instruction transmission interface: the ICE instruction transmission interface is an interactive interface packaged for efficiently transmitting the formation task instructions and the single-loading action instructions, and supports instruction transmission between different layers of levels and instruction transmission between a task planning intelligent agent and a simulation deduction environment by applying an ICE protocol. The invention mainly researches a layered task framework and a layered task planning method, and does not deeply introduce interface packaging.
3.2 finite-state-machine-based battle-class mission planning method design
3.2.1 mission task decomposition
The mission task is the embodiment of the intention and the result of a commander, is one of the input conditions for planning the mission task, and the decomposition of the mission task needs to be combined with elements such as battlefield environment, combat resources, mission task targets, combat phases and the like. The invention takes the fighting and fighting thought of the island reef as a sample case, and the thought content is as follows:
(1) battle resources:
TABLE 1 battle resource List
(2) The method comprises the following steps:
the scheme is that the red party (our party) defends the sea island and protects the radar building of our party as a target, and available combat units of our party mainly comprise fighters, early warning machines, naval vessels, submarines and missile launching vehicles. The blue side (enemy) can attack our side through the modes of air (main), water surface, underwater and the like, and the our side needs to defend the sea island within the specified time (6 hours), so that the radar building is not damaged.
According to mission task targets and combat resources, the tasks can be decomposed into regional air defense tasks, early warning reconnaissance tasks, maritime patrol tasks, interception and attack tasks, high-price aerial asset protection, withdrawal defense, return voyage supply and the like.
The regional air defense task is a battle-level task for defending against an important airspace by a fire network by sending equipment such as fighters, ships, ground air defense weapons and the like, and aims to prevent enemies from carrying out air defense.
The early warning reconnaissance mission is a mission for assigning a fighter plane and an early warning plane to reconnaissance and detect in a designated flight line or area, and aims to provide detection information and data for own parties.
The maritime patrol task is a task of assigning a protective ship, a submarine and the like to carry out battle patrol in a designated navigation line or area, and aims to provide detection capability support and firepower support and prevent an enemy from defending from the sea.
The interception and striking task refers to a task of sending combat airplanes, ships and ground air defense facilities to intercept and strike local equipment for penetration defense.
The high-value aerial asset protection task is a task of assigning a fighter to protect high-value targets such as an early warning aircraft and the like.
The withdrawal defense is a task that a unit of our party withdraws a security area to carry out defense combat under the disadvantage.
The return voyage supply refers to a task of returning to a base for maintenance and supply under the condition that the ammunition amount and the oil amount are insufficient or damaged.
3.2.2 formation resource allocation and force Generation
The formation resource allocation means that the battle resources are reasonably divided into different combinations to complete adaptive tasks, and the formation resource allocation is an optimization problem, but is not the focus of the research of the invention, so the invention adopts a manual grouping mode to divide the battle resources in the wanted mode into 12 groups, as shown in the following table 2:
TABLE 2 force marshalling List
The force generation refers to the fact that combat marshalling is constructed into Agent objects, and due to the fact that each Agent has different functions, tasks which can be executed by the agents are different; therefore, different Agent objects need to be constructed, 6 types of Agent objects are constructed according to the fighting scenario, namely a fighter fleet Agent, an early warning fleet Agent, a protective vessel fleet Agent, a submarine fleet Agent, a ground missile launching vehicle Agent and a radar Agent; each type of Agent is composed of static attribute information and a functional module, and the following table 3 shows the specific structure of the fighter team agents:
TABLE 3 Agent Structure
3.2.3 finite State machine based campaign-level mission planning
The operation planning is a dynamic process of condition judgment, behavior planning and adjustment, and due to the characteristics of unitization, intellectualization and the like of modern war, the problem of operation task planning is difficult to solve by using a mathematical modeling mode, so that military knowledge and expert experience are strongly required as a basis for decision making. Common knowledge management and organization methods include hard-coding-based methods, finite-state-machine-based methods, rule-engine-based methods, and knowledge-graph-based methods. The invention plans the battle-level tasks, and adopts a Finite State Machine (FSM) to organize the rules and generate the formation task instructions. Finite state machines are a class of mathematical models that describe state transitions within a system. Behaviors can be discretized into a finite number of states, and when a certain condition is met, the states of the behaviors jump.
The behavioral model described by the finite state machine can be expressed as:
FSM＝<StateSet,ConditionSet,f:(s,c),InitState,ActionSet>
wherein StateSet is a set of states of a behavior; conditionset is a condition set for state transition; f, (s, c) → s 'is a state transition function, which indicates that the current state s belongs to StateSet, and if the condition c belongs to ConditionSet is met, the state transition is s' belongedto StateSet; InitState represents the initial state of behavior; ActionSet is the set of actions that can be performed in each state.
The finite-state-machine-based campaign-level mission planning is developed for all agents, and at the same time, mission decisions are performed on all agents in parallel according to battlefield situation conditions, and finally, mission decision results of all agents are combined to form a joint mission sequence as output of a campaign-level mission planning layer, and fig. 2 shows a decision mechanism of the campaign-level mission planning layer.
The mission planning of the campaign layer has the following characteristics:
(1) mission tasks can be decomposed into limited subtasks according to time sequence and task purpose;
(2) when some subtasks are performed, the current situation needs to be analyzed and judged at any time, and the subtasks to be performed are changed at any time according to the situation.
(3) Some subtasks require coordination of other formations to be performed.
According to the above characteristics, it can be represented as follows by a simplified finite state machine:
Transitions＝(source,trigger,dest,after)
the source represents a task set which is currently executed, the task set is called as a state (state) in a state machine, trigger represents a trigger condition, the trigger is a task change judgment condition formed according to a situation, dest represents a next state, namely a task which needs to be executed after the trigger condition is met, and after is a series of operations which need to be executed for executing the task; transitions store the current state, change condition, and next state, which is another representation of the production rule IF-then structure. It is this structure that enables military knowledge to be used efficiently and reasonably in mission planning.
The process of planning the formation task using the finite state machine will be described below by taking the fighter formation with one initial task as the regional air defense task as an example. It is assumed that the formation of warplanes can perform 5 types of tasks when performing island and reef defense: regional air defense, airspace interception, high-value target attack, return voyage and retreat defense.
Firstly, setting an initial task of the fighter team as an area air defense task, and converting the initial task into a high-value target hitting task when an enemy ship is found; and when an enemy early warning machine or an interference machine is found, converting into an airspace interception task. And when the formation finishes striking or interception, the formation can be transferred to the regional air defense task again. When the oil quantity or the bomb quantity of the formation is insufficient, no matter what state the formation is in, the formation is converted into a return flight state, and a return flight instruction is executed.
Table 4 and fig. 3 show a state transition table and a state transition diagram of the fighter team, respectively.
TABLE 4 finite State of the fighter plane formation area air defense behavior described by State transition Table
Both the state transition table and the state transition diagram can be used as the representation form of the rule. For example, 5 rules are stored in a state conversion table for preventing air in a fighter plane formation area, wherein the rules are as follows:
(1) discovering enemy ships, namely performing high-value target striking rules;
(2) discovering an enemy early warning machine or an interference machine, namely carrying out airspace interception rules;
(3) after the high-value target is hit, executing the regional air defense rule;
(4) after the airspace interception is finished, executing the regional air defense rule;
(5) and when the formation oil quantity is insufficient or the bullet quantity is insufficient, the return flight is regulated.
3.2.4 formation task instruction Generation
The method can be used for obtaining the battle-level tasks of each formation at a dynamic moment, and the battle-level tasks need to be represented in a structured form in order to be transmitted to a tactical layer; wherein the formation campaign-level task instruction set is represented in the form:
CampaignSeqList＝[CampaignSeq1,CampaignSeq2,…,CampaignSeqN]
the specific task instruction CampaignSeqi includes the contents of task execution formation, task name, task type, task execution area and task target, and can be expressed in the following form:
CampaignSeq＝{Group,MissionName,MissionType,AreaVertex,Targets} i＝1,…,N。
3.3 tactical mission planning method design based on rules
3.3.1 formation task instruction resolution
And the battle mission planning layer encapsulates the output result, namely the battle mission instruction set, into a dictionary format and transmits the dictionary format to the tactic mission planning layer through an ICE protocol. To perform tactical task planning, the first task is to analyze an instruction set, and the steps are data decoding and data extraction, so that the formation tasks are extracted one by one and input into a tactical task planning layer.
3.3.2 hard-coded tactical layer mission planning
The tactical mission planning layer is a concrete plan for the battle mission, can be understood as that the mission of an intelligent agent of the tactical mission is a mission instruction output by the tactical mission planning layer, and is divided into 6 channels: planning regional air defense tasks, planning early warning and reconnaissance tasks, planning high-value aerial asset protection tasks, planning maritime patrol tasks, planning interception and attack tasks, planning formation return-voyage tasks, and synchronously planning tactical tasks of each formation. Fig. 4is a diagram of a hard-coded tactical layer mission planning architecture.
Different rules are needed for different tasks, but the tasks are basically organized by using an IF-then rule, and specific processes of the tasks are described by taking dual-machine area air defense tasks as a column. The main flow is shown in fig. 5.
After the task starts, the processes of battlefield situation acquisition, air combat decision, entity action output and the like are sequentially carried out, wherein the air combat decision is an intermediate link of the air defense task in the whole area and is also a key link, and situation evaluation analysis is the premise of the air combat decision. When the battlefield situation is evaluated and the air combat decision is made, the IF-THEN rule (shown in a judgment box in the figure) is mainly used, and the rules used in the regional air defense mission are as follows:
(1) if the fighter team of the fighter plane is not in the air defense area or the airplane of the enemy is not detected, the fighter team of the fighter plane is maneuvered into the air defense area and patrolling and detecting are carried out;
(2) if the fighter team of the party is in the air defense area and the enemy plane is detected, but the enemy plane is not found to enter the early warning area of the party, the fighter team of the party executes the tracking and monitoring task but cannot leave the early warning area;
(3) if I put the fighter team in the air defense area, scout the enemy plane and find that the enemy plane enters the early warning area of the party, the enemy plane is judged to have an attempt to attack the I, the fighter team executes the air-air combat task and intercepts and attacks the attacking enemy plane.
The output results of the tactical mission planning based on hard coding are formation tactical missions, such as formation area search, formation tracking enemy plane, formation take-off, formation maneuver, formation air-to-air combat and the like, which are respectively packaged as independent functions. Each function can refer to a combat rule, military knowledge and a mathematical planning method, and a multi-target distribution algorithm based on a genetic algorithm can be referred to in air-space combat.
3.3.3 Single load instruction Generation
The output of the tactical mission planning layer is a single-loading instruction list, the single-loading instruction is an instruction which can be directly applied to a platform, such as an air route maneuvering instruction, a take-off instruction, an air-to-air missile launching instruction, a sensor opening instruction and the like, and each instruction is the decomposition of a formation tactical mission. Because the method is simple and is not described in detail, the final result is that a single-loading task instruction sequence is generated and transmitted to the simulation deduction platform through an ICE code. The instruction sequence structure is as follows:
ActionSeqList＝[ActionSeq1,ActionSeq2,…,ActionSeqN]
the single one-piece task instruction ActionSeqi includes an action entity, an action name, an action type, a target, and an action route, which can be expressed as the following form:
ActionSeq＝{Entity,ActionName,ActionType,Target,ActionRoute}。
the above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (10)
1. A rule-based sea-air combined combat action layered task planning method is characterized in that a sea-air combat process for island reef attack and defense planning is decomposed into a high-level battle-level combat process and a low-level tactic-level combat process according to an OODA (object oriented data acquisition) ring model;
the high-level battle-level combat process is represented by an OODA (object oriented data acquisition) ring with a long period, and means that battle field situations are judged by collecting battle field information under the constraint of a specific mission task and combat capability, so that the planning of the battle tasks and the issuing of combat instructions are realized;
the low-level tactical combat process is represented by using an OODA ring with a short period, and means that on the basis of evaluating the combat capability and threat situation according to an instruction issued by a superior level, the local battlefield information is collected and the local battlefield situation is judged, so that the mission planning and tactical decision of basic combat marshalling or single-package basic action are realized; the low-level tactical combat process runs in an OODA ring of the high-level tactical combat process, the OODA ring of each low-level tactical combat process represents a subtask, the execution result of all the subtasks changes the battlefield situation to influence the high-level tactical combat process, and judgment and decision are taken as the content of command and decision activities and are carried out in a rule reasoning mode.
2. The method according to claim 1, wherein the method constructs a rule-based hierarchical mission planning model structure to describe the sea-air combat process for island-reef defense and defense planning according to the hierarchical characteristics of the sea-air combat process, the model structure comprises two levels of a battle mission planning model and a tactic mission planning model, the two levels respectively correspond to a high-level battle-level combat process and a low-level tactic-level combat process, and the model structure provides a situation awareness interface and a command transmission interface for acquiring battle field situation data and issuing mission commands respectively.
3. The method of claim 2, wherein the campaign mission planning layer is designed to: reflecting the process of carrying out operation planning according to the operation purpose and mission task of a commander under the global situation, and the realized specific planning flow comprises the following stages: mission task decomposition, formation resource distribution and force generation, finite-state machine-based battle layer mission planning and formation mission instruction generation;
the tactical mission planning layer is designed as follows: reflecting the collaborative planning of the battle formation, solidifying the relevant knowledge and collaborative method of tactical planning in the bottom planning model of the tactical mission planning layer, and forming the corresponding planning result on the basis of the battlefield situation and the upper-layer instruction; the realized specific planning process comprises the following stages: and analyzing the formation task instruction, planning the tactical layer task based on hard coding, and generating a single-package instruction.
4. A method as claimed in claim 3, characterized in that in the method the situational awareness interface is designed to: the layered battle planning intelligent body provides battlefield situation data and is packaged with an interactive interface of the environment, and detection data, war newspaper data, battle damage data and equipment state data can be extracted from a battle simulation environment; the instruction transmission interface is designed as an ICE instruction transmission interface, the ICE instruction transmission interface is an interactive interface packaged for transmitting the formation task instruction and the single-loading action instruction, and an ICE protocol is applied to support instruction transmission between different levels and instruction transmission between a task planning intelligent agent and a simulation deduction environment.
5. The method of claim 3, wherein the mission resolution stage in the planning process of the campaign mission planning layer is designed to: according to mission task targets and combat resources, tasks are decomposed into regional air defense tasks, early warning reconnaissance tasks, maritime patrol tasks, interception and attack tasks, high-price aerial asset protection, withdrawal defense and return voyage supply;
the regional air defense task is a battle-level task for assigning equipment such as fighters, ships and ground air defense weapons to form a fire power network to defend a preset key airspace, and aims to prevent enemies from performing air defense; the early warning reconnaissance task is a task of assigning a fighter plane and an early warning plane to perform reconnaissance detection in a specified airline or area, and aims to provide detection information and data for own parties; the maritime patrol task is a task of dispatching a guard ship and carrying out battle patrol on equipment in a designated navigation line or area by a submarine, and aims to provide detection capability support and firepower support and prevent an enemy from defending from the sea; the interception and striking task is a task of sending a combat aircraft, a ship and ground air defense facilities to intercept and strike local equipment for penetration defense; the high-value aerial asset protection task is a task of assigning a fighter to protect a high-value target, namely an early warning aircraft; withdrawal defense is a task that under the disadvantage, the unit of our party withdraws a security area to carry out defense combat; the return voyage supply refers to a task of returning to a base for maintenance and supply under the condition that the ammunition amount and the oil amount are insufficient or damaged.
6. The method of claim 5, wherein the allocation of queuing resources and the generation of power in the planning process of the campaign mission planning layer are designed to:
and (3) queuing resource allocation: dividing the combat resources into different combinations for completing adaptive tasks; adopting a manual marshalling mode to divide combat resources which want to be centered into 12 marshalling;
force generation: constructing a battle marshalling into an Agent object, wherein each Agent can execute different tasks; constructing different Agent objects, and obviously constructing 6 types of Agent objects according to combat scenarios; each type of Agent is composed of static attribute information and a functional module.
7. The method of claim 6, wherein the stages of finite state machine based campaign-level mission planning in the planning flow of the campaign mission planning layer are designed to:
planning the operation: for a dynamic condition judgment, behavior planning and adjustment process, a finite state machine FSM is adopted to carry out regular organization and formation task instruction generation;
the finite-state-machine-based battle-level mission planning is developed for all Agent objects, and at the same time, according to battlefield situation conditions, mission decisions are performed on all Agent objects in parallel, and finally, mission decision results of all agents are combined to form a joint mission sequence as output of a battle-level mission planning layer;
according to the characteristics of the mission plan of the campaign layer, the mission plan of the campaign layer is represented by a simplified finite-state machine as follows:
Transitions＝(source,trigger,dest,after)
the source represents a task set which is currently executed, the task set is called as a state (state) in a state machine, trigger represents a trigger condition, the trigger is a task change judgment condition formed according to a situation, dest represents a next state, namely a task which needs to be executed after the trigger condition is met, and after is a series of operations which need to be executed for executing the task; transitiond stores the current state, change condition, and next state, which is another representation of the production rule IF-then structure.
8. The method of claim 7, wherein the generation phase of the formation task instructions in the planning flow of the campaign mission planning layer is designed to:
after the battle-level tasks of the formation at the dynamic moment are obtained, the instruction set of the battle-level tasks of the formation is expressed as the following form:
CampaignSeqList＝[CampaignSeq1,CampaignSeq2,…,CampaignSeqN]
the specific task instruction CampiagnSeqi comprises the contents of task execution formation, task name, task type, task execution area and task target;
the output result, i.e., the set of tactical mission instructions, may then be packaged into a dictionary format for delivery to the tactical mission planning layer via the ICE protocol.
9. The method of claim 8, wherein the team task instruction parsing stage in the planning process of the tactical mission planning layer is designed to: and through data decoding and data extraction and analysis of a battle task instruction set, the formation tasks are extracted one by one and input into a tactic task planning layer.
10. The method of claim 9, wherein the hard-coded tactical layer mission planning phase of the tactical mission planning layer planning flow is designed to:
the tactical mission planning layer is divided into 6 channels: planning regional air defense tasks, planning early warning and reconnaissance tasks, planning high-value aerial asset protection tasks, planning maritime patrol tasks, planning interception and attack tasks, planning formation return-voyage tasks, and synchronously planning tactical tasks of each formation;
after the task starts, battlefield situation acquisition, air combat decision and entity action output flow are carried out in sequence, situation evaluation analysis is firstly carried out before the air combat decision, the battlefield situation is evaluated and the air combat decision is carried out by using an IF-THEN rule, and the rules used in the regional air defense task are as follows:
(1) if the fighter team of the fighter plane is not in the air defense area or the airplane of the enemy is not detected, the fighter team of the fighter plane is maneuvered into the air defense area and patrolling and detecting are carried out;
(2) if the fighter team of the party is in the air defense area and the enemy plane is detected, but the enemy plane is not found to enter the early warning area of the party, the fighter team of the party executes the tracking and monitoring task but cannot leave the early warning area;
(3) if I put the fighter team in the air defense area, scout the enemy plane and find that the enemy plane enters the early warning area of the party, then judge that the enemy plane has an attempt to attack the I, and I fighter team executes the air-air combat mission and intercepts and attacks the attacking enemy plane;
the output result of the tactical mission planning based on hard coding is a formation tactical mission, which comprises formation area search, formation tracking enemy plane, formation take-off, formation maneuvering and formation air-to-air combat, which are respectively packaged as independent functions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111630161.5A CN114282833A (en) | 2021-12-28 | 2021-12-28 | Rules-based hierarchical task planning method for air-sea combined combat action |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111630161.5A CN114282833A (en) | 2021-12-28 | 2021-12-28 | Rules-based hierarchical task planning method for air-sea combined combat action |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114282833A true CN114282833A (en) | 2022-04-05 |
Family
ID=80877350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111630161.5A Pending CN114282833A (en) | 2021-12-28 | 2021-12-28 | Rules-based hierarchical task planning method for air-sea combined combat action |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114282833A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114897267A (en) * | 2022-06-14 | 2022-08-12 | 哈尔滨工业大学（深圳） | Fire power distribution method and system for many-to-many intelligent agent cooperative battlefield scene |
| CN115017332A (en) * | 2022-06-07 | 2022-09-06 | 北京庚图科技有限公司 | Method and system for generating combat instruction |
| CN115357051A (en) * | 2022-10-18 | 2022-11-18 | 北京理工大学 | Deformation and maneuvering integrated avoidance and defense method |
| CN116151492A (en) * | 2023-04-21 | 2023-05-23 | 中国兵器科学研究院 | Auxiliary decision analysis system of combined combat system |
| CN117634858A (en) * | 2024-01-26 | 2024-03-01 | 中国人民解放军军事科学院国防科技创新研究院 | Man-machine mixing scheme generation method and device based on instant advantage window |
-
2021
- 2021-12-28 CN CN202111630161.5A patent/CN114282833A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115017332A (en) * | 2022-06-07 | 2022-09-06 | 北京庚图科技有限公司 | Method and system for generating combat instruction |
| CN114897267A (en) * | 2022-06-14 | 2022-08-12 | 哈尔滨工业大学（深圳） | Fire power distribution method and system for many-to-many intelligent agent cooperative battlefield scene |
| CN114897267B (en) * | 2022-06-14 | 2024-02-27 | 哈尔滨工业大学（深圳） | Fire distribution method and system for multi-to-multi-agent cooperative combat scene |
| CN115357051A (en) * | 2022-10-18 | 2022-11-18 | 北京理工大学 | Deformation and maneuvering integrated avoidance and defense method |
| CN115357051B (en) * | 2022-10-18 | 2023-02-03 | 北京理工大学 | Deformation and maneuvering integrated avoidance and defense method |
| CN116151492A (en) * | 2023-04-21 | 2023-05-23 | 中国兵器科学研究院 | Auxiliary decision analysis system of combined combat system |
| CN117634858A (en) * | 2024-01-26 | 2024-03-01 | 中国人民解放军军事科学院国防科技创新研究院 | Man-machine mixing scheme generation method and device based on instant advantage window |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114282833A (en) | Rules-based hierarchical task planning method for air-sea combined combat action | |
| Mour et al. | Agent‐Based modeling for systems of systems | |
| Biltgen et al. | A Methodology for Capability-Focused Technology Evaluation of Systems of Systems | |
| Barry et al. | UCAVs, Technological, Policy, and Operational Challenges | |
| Grecu et al. | Agent-based simulation environment for UCAV mission planning and execution | |
| Chapman II | Unmanned combat aerial vehicles: Dawn of a new age? | |
| CN116502423A (en) | Counter-simulation deduction system based on deep reinforcement learning | |
| McLemore et al. | A model for geographically distributed combat interactions of swarming naval and air forces | |
| Huang et al. | Research on Intelligent Cooperative Command and Control System of Unmanned Equipment in Amphibious Operations | |
| Lotter | Design of a weapon assignment subsystem within a ground-based air defence environment | |
| Duan et al. | Evaluation Method of UAV’s Contribution Degree to Army Aviation Combat System Based on MMF and ANP | |
| Ghigliotti et al. | Artificial Intelligence-Enabled Multi-Mission Resource Allocation Tactical Decision Aid | |
| Blandin et al. | 2024 Unmanned undersea warfare concept | |
| Ghigliotti et al. | SYSTEMS ENGINEERING CAPSTONE REPORT | |
| Shi et al. | Modeling Technology of Naval Formation Area Antiaircraft Kill Web based on Event Graph | |
| Ren et al. | Swarm Operation of Heterogeneous Electronic Countermeasure UAV in Denial Environment | |
| Guo et al. | Research on optimal allocation of surface ship formation based on index method and multi-objective programming | |
| Buss et al. | Distributed surface force | |
| FORCE | SYSTEMS ENGINEERING ANALYSIS CAPSTONE PROJECT REPORT | |
| Shi et al. | Knowledge Representation and Reasoning of Strong System Confrontation Process Based on Spatio-Temporal Event Evolution Information Flow and Software | |
| Yan et al. | Index System for Autonomous Capability of Ground-Attack UAV | |
| Seres | The Armed Combat as a System | |
| Letourneau | Incorporating multi-criteria optimization and uncertainty analysis in the model-based systems engineering of an autonomous surface craft | |
| Jin et al. | Research on Hybrid Intelligence Wargame Method | |
| Guo et al. | Research on Simulation Design of Air Force Campaign Process Deduction System |
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 |