CN114792072B - Function-based equipment decision behavior simulation modeling method and system - Google Patents
Function-based equipment decision behavior simulation modeling method and system Download PDFInfo
- Publication number
- CN114792072B CN114792072B CN202210551852.4A CN202210551852A CN114792072B CN 114792072 B CN114792072 B CN 114792072B CN 202210551852 A CN202210551852 A CN 202210551852A CN 114792072 B CN114792072 B CN 114792072B
- Authority
- CN
- China
- Prior art keywords
- decision
- node
- nodes
- edges
- output
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000005094 computer simulation Methods 0.000 title claims abstract description 15
- 238000004458 analytical method Methods 0.000 claims abstract description 64
- 230000009471 action Effects 0.000 claims abstract description 49
- 238000003066 decision tree Methods 0.000 claims abstract description 37
- 238000004088 simulation Methods 0.000 claims abstract description 33
- 230000000875 corresponding effect Effects 0.000 claims abstract description 27
- 238000004064 recycling Methods 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 4
- 238000004880 explosion Methods 0.000 abstract description 3
- 230000006399 behavior Effects 0.000 description 30
- 230000002068 genetic effect Effects 0.000 description 22
- 238000010586 diagram Methods 0.000 description 14
- 230000006870 function Effects 0.000 description 13
- 230000008569 process Effects 0.000 description 12
- 238000004422 calculation algorithm Methods 0.000 description 8
- 230000007123 defense Effects 0.000 description 8
- 238000011156 evaluation Methods 0.000 description 6
- 238000013461 design Methods 0.000 description 4
- 210000000349 chromosome Anatomy 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 210000004556 brain Anatomy 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000010304 firing Methods 0.000 description 2
- 239000011888 foil Substances 0.000 description 2
- 239000010763 heavy fuel oil Substances 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 238000000342 Monte Carlo simulation Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000001010 compromised effect Effects 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000005669 field effect Effects 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012905 input function Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000035800 maturation Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/02—Computing arrangements based on specific mathematical models using fuzzy logic
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Biophysics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Algebra (AREA)
- Automation & Control Theory (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Computational Linguistics (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a functional-based equipment decision behavior simulation modeling method and system. The method comprises the following steps: s1, constructing a functional decision tree, wherein the functional decision tree comprises a meta model root node, the meta model root node comprises tree nodes and edges, the tree nodes comprise situation analysis nodes, decision nodes and action nodes, the edges comprise data edges for transmitting data and decision edges for decision control, and the situation analysis nodes, the decision nodes and the action nodes are associated through the data edges and the decision edges; s2, simulation is executed, the node type is judged, corresponding decision is executed if the node is a decision node, the next decision node is returned, iteration is carried out if the next decision node is not empty, corresponding action is directly executed if the node is an action node, topology ordering is firstly carried out on all the nodes if the node is a situation analysis node, and each situation analysis node in ordering is executed in a recycling mode to obtain a final output value. The invention can effectively relieve the problem of state space combination explosion caused by excessive input variables.
Description
Technical Field
The invention relates to the technical field of decision making, in particular to a functional equipment decision behavior simulation modeling method and system.
Background
The fighter efficiency evaluation is an important content supporting the full life cycle work of fighter equipment demonstration, development, use and the like, and the fighter simulation has become a main technology of fighter efficiency evaluation along with the maturation of simulation technology and the improvement of efficiency evaluation requirements. The fighter efficiency simulation evaluation is not supported by a fighter simulation system with high efficiency and reliability, and the decision modeling of a pilot belongs to a key ring of the fighter simulation system construction. The existing fighter fight efficiency evaluation fight simulation system generally lacks an effective air fight decision modeling method, and mostly adopts reactive rules to carry out scripted modeling, so that the simulated air fight decision behavior has large difference from the actual fight, the predictability is strong, the reliability of the fighter fight simulation effect is influenced due to lack of deliberations, and the play of the fighter efficiency evaluation field effect is restricted.
In 2016, the artificial intelligence system Alpha (Alpha AI) of air combat developed by the university of Xincinnati in the United states and the air force in the United states has defeated the retirement of the air force in a simulated air combat, and this event has attracted widespread attention both at home and abroad. The alpha adopts an air combat decision simulation modeling method based on a genetic fuzzy tree, a plurality of input and output variables of a decision space are subjected to fuzzification or classification, air combat decision fuzzy reasoning rules are generated based on a genetic algorithm, and are connected in a tree structure cascade manner, so that the human brain can be effectively simulated to implement rule type reasoning to implement upper layer decision intention, and credible behaviors are generated. However, for some data with a well-defined binary meaning, or discrete values with well-defined semantics, blurring using membership functions is not facilitated. For this situation, one way of naive approach is to design a simple classifier for the fuzzy inference output, and discretize the continuous values to correspond to the original discrete outputs respectively. It is clear that this tends to be highly subjective, and model interpretability is rather compromised.
The current typical air combat decision behavior modeling methods at home and abroad are mainly three types: knowledge engineering, fuzzy logic, and machine learning, wherein fuzzy logic is often used in conjunction with heuristic intelligent optimization algorithms. The first two are mainly analyzed by taking a state diagram and a genetic fuzzy tree as examples.
Fighter states in air combat generally undergo basic stages of take-off, cruising, detection, reception, engagement, termination of mission, return to the base. After the fighter receives the fighter command, the fighter takes off and flies along a preset route at a certain cruising speed, and meanwhile, the airborne radar is started to search for a target. When an attack target is found and identified, a receiving stage is entered, and the target is maneuvered and occupied according to the position and the speed of the target. When the weapon firing condition is reached, the weapon is fired to enter the fight stage, and if the enemy is found to attack the missile, the fighter plane gets rid of rapidly. The mission is over or the fuel is exhausted and the aircraft returns to the base.
According to the air combat characteristics and the state diagram modeling specifications, the fighter process can be abstracted into a series of states and complex state transition relations among the states. Practice shows that the air combat decision behavior modeling based on the state diagram has certain limitation when describing the air combat decision behavior, and cannot truly reflect the actual air combat process, and is mainly represented by: firstly, the state transition relation is complex and is not easy to maintain. Future air combat is combined and integrated cross-domain combat, the contained weapons are numerous, tactical tactics are complex, describing the weapons by adopting a state diagram can lead to a large number of states and complex transfer relations among the states, the maintenance is not easy, the expansibility is poor, and the addition of one state often causes the change of the whole state diagram, so that a large number of debugging works are caused. Secondly, the expression capability is limited, and uncertainty expert knowledge cannot be described. The state diagram can only describe deterministic expert knowledge, but cannot describe uncertain expert experience. With the superposition of battlefield dimensions, the innovation of weapon equipment and tactical battle law inevitably generates more complex and large amount of uncertain battlefield information, and the uncertain information of the battlefield must be described to truly reflect the actual battlefield process. Thirdly, the modeling difficulty is high by a non-modularized modeling method. The state diagram adopts a non-modularized modeling method, so that the regulations are not clear enough, modeling difficulty is increased, modeling results cannot be effectively reserved, communication and compatibility between models are inconvenient, new modeling staff cannot participate in modeling work quickly, a large amount of unnecessary workload is caused, and the efficiency of behavior modeling is reduced.
The evolution process of the genetic fuzzy tree (Genetic Fuzzy Tree, GFT) mainly goes through several processes of genetic algorithm (Genetic Algorithm, GA), fuzzy inference system (Fuzzy Inference System, FIS), genetic fuzzy system (Genetic Fuzzy System, GFS), genetic cascade fuzzy system (Genetic Cascading Fuzzy System, GCFS) and genetic fuzzy tree. On one hand, the genetic fuzzy tree is based on fuzzy logic and adopts a cascading tree structure, so that uncertainty information can be effectively expressed, meanwhile, the complexity of the problem can be effectively reduced through a modularized structure, and the readability and the understandability of the model are improved. Fuzzy logic is good at expressing qualitative knowledge and experience with unclear boundaries, and simulates human brain to implement rule type reasoning by means of membership function concept, so that the problem of rule type fuzzy information which is difficult to cope with by conventional methods is solved by reasoning. On the other hand, the genetic fuzzy tree encodes the related parameters of the rule set and the membership function into chromosomes by using a genetic algorithm, and then uses an iterative mode to perform operations such as selection, crossover, mutation and the like to exchange the information of the chromosomes in the population, so as to finally generate the chromosomes [ i ] [ ii ] which meet the optimization target. However, genetic fuzzy trees still have the following problems: firstly, decision making reasoning and behavior actions are not clearly distinguished, so that a model is confused. In practice, the decision command output belongs to the action of the airplane behavior, is generated from decision reasoning and affects the decision, and has no intersection in the decision reasoning process, but directly acts on the airplane after the current decision command is generated. Without distinction during the modeling process, difficulties can be presented to understanding and maintaining the model. And secondly, the fuzzy inference system has continuous output and limited expression capability. The genetic fuzzy tree infers different levels of decision variables, and the defuzzified value is a continuous value as the output of the system. However, in air combat decision-making, there are discrete decision points of input and output, such as whether to launch a missile, whether to turn on a radar, weapon type selection, and the like. When the fuzzy inference system is used for description, the discretization meaning of the continuous output value is poor in interpretation. Thirdly, the model is suitable for specific decision problems and has poor compatibility and inheritance. The fuzzy decision tree model established according to the existing rules is only suitable for solving the specific decision problem. If the combat scene changes, equipment is innovated, the existing decision model cannot be inherited, the modeling difficulty is increased again, the model expansibility is poor, other methods are difficult to be compatible, and the compatibility is poor.
Disclosure of Invention
The invention aims to provide a functional equipment decision behavior simulation modeling method and system, which are used for overcoming the defects existing in the prior art.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
a functional equipment decision behavior simulation modeling method comprises the following steps:
s1, constructing a functional decision tree, wherein the functional decision tree comprises a meta-model root node, the meta-model root node comprises tree nodes and edges, the tree nodes comprise situation analysis nodes, decision nodes and action nodes, the edges comprise data edges for transmitting data and decision edges for decision control, the situation analysis nodes, the decision nodes and the action nodes are related through the data edges and the decision edges, the situation analysis nodes execute corresponding calculation according to input situation data and output data, the decision nodes input the output data of the situation analysis nodes to make internal decisions and return to the next decision node needing to be decided, and the action nodes execute related actions according to decision results;
s2, performing simulation, judging the node type, performing corresponding decision and returning to the next decision node if the node type is the decision node, iterating if the next decision node is not null, directly performing corresponding action if the node type is the action node, and performing topological sorting on all nodes if the node type is the situation analysis node, and performing each situation analysis node in sorting in a recycling manner to obtain a final output value.
Further, the situation analysis node has 0 or more output data edges, the action node has and has only one output decision edge, the decision node has and has only one input decision edge, the decision node has 0 or more output decision edges, and the decision node receives the data output by the situation analysis node and has 0 or more input data edges.
Further, the step of performing the topological ordering on all the nodes in the step S2 specifically includes: initializing a sorting node list, acquiring all nodes with output, acquiring all nodes with input, calculating all node lists with output nodes but without input, calculating all node lists with input nodes but without output, returning to the sorting node list if all node lists with output nodes but without input are not empty, and outputting the sorted sorting node list.
Further, the functional decision tree exists in the form of a Python script, and each element in the functional decision tree metamodel is mapped into a corresponding script element.
The invention also provides a system of the functional equipment decision behavior simulation modeling method, which comprises:
the construction module is used for constructing a functional decision tree, the functional decision tree comprises a meta-model root node, the meta-model root node comprises tree nodes and edges, the tree nodes comprise situation analysis nodes, decision nodes and action nodes, the edges comprise data edges for transmitting data and decision edges for decision control, the situation analysis nodes, the decision nodes and the action nodes are related through the data edges and the decision edges, the situation analysis nodes execute corresponding calculation according to input situation data and output data, the decision nodes input the output data of the situation analysis nodes to make internal decisions and return to the next decision node needing to be decided, and the action nodes execute related actions according to decision results;
the simulation execution module is used for executing simulation, judging the node type, executing corresponding decision and returning to the next decision node if the node type is the decision node, iterating if the next decision node is not empty, directly executing corresponding action if the node type is the action node, and performing topology sequencing on all the nodes if the node type is the situation analysis node, and performing each situation analysis node in sequencing in a recycling mode to obtain a final output value.
Compared with the prior art, the invention has the advantages that: the invention adopts the function decision tree, which is to define the behavior model from the view of functions, simulation application personnel does not need to define and maintain a large number of variables such as states, events and the like, and can effectively relieve the problem of state space combination explosion caused by excessive input variables.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a diagram of a functional decision tree metamodel in the present invention.
Fig. 2 is a diagram of a situation analysis process in the present invention.
Fig. 3 is a diagram of an attack decision process in the present invention.
Figure 4 is a diagram of a defensive decision process in the present invention.
FIG. 5 is a graph showing the output of the Alpha simulation results of the fighter plane in the present invention.
FIG. 6 is a two-dimensional display diagram of a single air combat simulation experiment in the invention.
Detailed Description
The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making clear and defining the scope of the present invention.
The design concept of the invention is a functional decision tree, and the functional decision tree is five-tuple FDT= (T, N) analysis ,N action ,N decision ζ, δ, λ), where T is the time set, N analysis To analyze node sets, N action N is the action node set decision For the decision node set, ζ: N analysis →N decision Data input function, delta: N decision ×N analysis →N decision For controlling functions for decision nodes, lambda: N decision ×N analysis →N action The function is output for the action.
The functional decision tree design concept mainly comprises the following three points: firstly, unlike the description modeling method based on states and events adopted by most combat simulation systems at present, the functional decision tree defines a behavior model from the functional perspective, simulation application personnel do not need to define and maintain a large number of variables such as states and events, and the problem of state space combination explosion caused by excessive input variables can be effectively relieved. Secondly, the functional decision tree adopts a tree-shaped hierarchical structure to describe node control relations between decisions and actions, and simulation modeling staff can conveniently delete, add, multiplex and the like nodes in the functional decision tree model, so that the functional decision tree is easy to maintain and expand. Thirdly, tree nodes in the functional decision tree are a highly abstract decision behavior, and can be output after being seen that a black box receives relevant input and carries out corresponding processing, modeling staff does not need to pay attention to internal calculation of the nodes, and the modeling method is expected to be compatible with other typical simulation modeling methods such as a behavior tree, a state machine, a genetic fuzzy tree, a neural network and the like.
Referring to fig. 1, the embodiment discloses a functional equipment decision behavior simulation modeling method, which comprises the following steps:
step S1, a functional decision Tree is built, the functional decision Tree comprises a meta model root node Tree, the meta model root node comprises Tree nodes and edges, the Tree nodes comprise situation analysis nodes, decision nodes and action nodes, the edges comprise data edges for transmitting data and decision edges for decision control, the situation analysis nodes, the decision nodes and the action nodes are related through the data edges and the decision edges, the situation analysis nodes execute corresponding calculation according to input situation data and output data, the decision nodes input the output data of the situation analysis nodes to make internal decisions and return to the next decision node needing to be decided, and the action nodes execute relevant actions according to decision results.
In this embodiment, the situation analysis node has 0 or more output data edges, the action node has and has only one output decision edge, the decision node has and has only one input decision edge, there are 0 or more output decision edges, and the decision node receives the data output by the situation analysis node, and there are 0 or more input data edges.
Step S2, executing simulation, judging the node type, executing corresponding decision if the node type is a decision node, returning to the next decision node, iterating if the next decision node is not empty, and directly executing corresponding action if the node type is an action node, wherein the simulation algorithm is as shown in the following table 1.
TABLE 1
If the situation analysis nodes are present, each situation analysis node in the tree generates output by processing the output and transmits the output to other nodes, and the input-output relationship adopts the idea of forward feedback. And firstly, carrying out topological sorting on all the nodes, and carrying out each situation analysis node in the sorting to obtain a final output value, as shown in a table 2.
TABLE 2
In this embodiment, the functional decision tree exists in the form of a Python script, and each element in the functional decision tree metamodel is mapped into a corresponding script element. The tree node is an abstract node and is inherited by a situation analysis node, a decision node Decisionnode and an action node Actionnode. Specific:
(1) And the situation analysis node inputs the combat situation data, performs corresponding calculation, preprocesses the data and outputs the data required by decision making to the decision making node. Mainly comprises three types of combat situation data: first, target state information such as target distance and target speed; secondly, the performance parameters of the aircraft, such as residual fuel oil and residual missile; and thirdly, fight situation data, such as whether the fight situation data is locked by an enemy radar, missile warning and the like. The AnalysisNode calculates through a member function of input_par (output_par), and all child nodes need to rewrite the function to execute corresponding calculation. Situation data, which typically does not perform any calculations, may also be directly input into the decision node.
(2) The decision node is essentially a control node and comprises an input edge and a plurality of output edges, the data obtained by the analysis node is input to make internal decisions, one of the plurality of output edges is selected, and the next node needing to be decided is returned. In the whole single-machine air combat process, each link needing decision is generally abstracted into a decision node, such as combat decision, attack, defense, course flight, return flight, weapon selection, weapon emission and the like. The decision node realizes internal decision logic by a user through a make_decision (input_par) function, and outputs the next decision node needing decision.
(3) Action nodes, which generally appear as leaf nodes, should inherit from an action node class and rewrite the do_action (input_par) function to perform the relevant actions. Such as for example, return voyage, launch of mid-long range air-to-air missiles, launch of near range air-to-air missiles, automatic avoidance missiles, launch of foil strips, launch of tow missiles, and the like.
The invention is further described below by way of specific examples.
The case is based on a weapon equipment effectiveness simulation system (Weapon Effectiveness Simulation System, WESS) that supports the development, operation and analysis of equipment combat simulation applications, including equipment data management, intended editing, decision modeling, simulation experiment design, monte carlo simulation operation, simulation experiment management, simulation data acquisition, two-dimensional/three-dimensional distributed representation, record playback, and the like. In this embodiment, a single air combat is taken as an example, and a decision modeling link is focused on.
The single machine air combat simulation experiment shows that the red and blue two systems are identical, and 4 long-distance air-to-air missiles and one airborne fire control radar are arranged in each system. The only difference is that the red Fighter (Fighter) uses a default state diagram based decision behavior script (cmaeroobject. Py), while the blue Fighter (Alpha) uses a functional decision tree based decision behavior script (cmalpha. Py) herein. After taking off, the two parties fly along the respective preset routes, the routes are intersected, and when one party enters the detection range of the other party, the two parties start to fight after being successfully identified. When any party is knocked down or reaches the last waypoint, the task is completed, and the simulation is ended.
The air combat decision behavior model based on the functional decision tree is shown in fig. 2-4, and is mainly divided into three parts, namely situation analysis, attack decision and defense decision.
(1) And (5) situation analysis. The situation awareness part applies analysis nodes, mainly represents information such as target state information, performance parameters of the aircraft, real-time combat situations and the like, and finally meets a situation decision node (status decision). The target state information includes a target distance (TargetDistance) and a target speed (TargetVelocity); aircraft own performance parameters include residual fuel (RemainingFuel) and residual missiles (remainingmessles); the operational situation includes whether locked by radar (IsLockByRadar) and whether missile warning (MissileWarning).
(2) Attack decision. Attack decisions mainly include beyond visual range combat (BVR) and close range combat (CAC). Beyond-the-horizon combat includes tracking (AutoTrace), and locking (LockOn) and launching of an medium-range air-to-air missile (launchfaram); close combat includes pursuit (autocase) and launch of close range air-to-air missiles (LaunchNearAAM).
(3) Defense decisions. The defense decisions mainly include both active defense (activedenfense) and passive defense (passedenfense) types. Active defenses include getting rid of lock (AutoBreakLock) and opening electronic interference measures (OpenECM); passive defenses include evading missiles (autoavoidmissille), throwing foil strips (LauchChaff), and launching a glowing projectile (LaunchFlare).
And after the simulation operation is finished, the Alpha simulation operation output of the blue fighter is shown in fig. 5. Alpha captures the target for the first time at 1278 and recognizes the target as an aircraft at 1448, where both red and blue are about 130 km away. And then, alpha scans a decision behavior model according to the situation data perceived in real time, makes a round of decisions, executes decision nodes, makes decision sequences of situation decisions, attacks and beyond-view combat, and launches a medium-distance air-to-air missile along the 11 o' clock direction. At 1658, the red aircraft Fighter is knocked down by the fourth medium-to-long air-to-air missile launched by the blue aircraft Alpha.
Simulation two-dimensional display is shown in fig. 6, it is obvious that the Fighter entity created by both red and blue aircraft, alpha, issues a weapon-firing instruction to fire the first medium-long distance air-to-air missile, and attacks the red aircraft filter at 1449.
In view of the complexity and uncertainty of the air combat decision behaviors, the traditional deterministic behavior modeling method based on the state diagram is difficult to truly reflect the real combat decision behaviors. The genetic fuzzy tree utilizes the fuzzy logic to process the fuzzy data of uncertainty in the air combat, the built model has good layering property, is easy to expand and maintain, and optimizes the combat behavior rules and membership functions by utilizing the genetic algorithm, thereby obtaining good effects. However, for some variables with definite discrete semantics, especially "yes/no" binary variables, if the fuzzy processing is performed by applying a genetic fuzzy tree, the decision quality of fighter plane behaviors is too cumbersome and can be affected. On the basis of researching a genetic fuzzy tree, the invention provides the air combat decision behavior modeling method based on the functional decision tree, and from the functional perspective, decision modeling staff only need to determine the input and output of each decision, does not need to pay attention to internal decision logic, and can have the characteristics of various simulation modeling methods. And the next step of working is to integrate machine learning intelligent algorithms such as a neural network, a support vector machine and the like, so as to further verify the effectiveness and universality of the functional decision tree in the field of air combat decision behavior modeling.
The invention also provides a system of the functional equipment decision behavior simulation modeling method, which comprises: the construction module is used for constructing a functional decision tree, the functional decision tree comprises a meta-model root node, the meta-model root node comprises tree nodes and edges, the tree nodes comprise situation analysis nodes, decision nodes and action nodes, the edges comprise data edges for transmitting data and decision edges for decision control, the situation analysis nodes, the decision nodes and the action nodes are related through the data edges and the decision edges, the situation analysis nodes execute corresponding calculation according to input situation data and output data, the decision nodes input the output data of the situation analysis nodes to make internal decisions and return to the next decision node needing to be decided, and the action nodes execute related actions according to decision results; the simulation execution module is used for executing simulation, judging the node type, executing corresponding decision and returning to the next decision node if the node type is the decision node, iterating if the next decision node is not empty, directly executing corresponding action if the node type is the action node, and performing topology sequencing on all the nodes if the node type is the situation analysis node, and performing each situation analysis node in sequencing in a recycling mode to obtain a final output value.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, the patentees may make various modifications or alterations within the scope of the appended claims, and are intended to be within the scope of the invention as described in the claims.
Claims (2)
1. The functional equipment decision behavior simulation modeling method is characterized by comprising the following steps of:
s1, constructing a functional decision tree, wherein the functional decision tree comprises a meta-model root node, the meta-model root node comprises tree nodes and edges, the tree nodes comprise situation analysis nodes, decision nodes and action nodes, the edges comprise data edges for transmitting data and decision edges for decision control, the situation analysis nodes, the decision nodes and the action nodes are related through the data edges and the decision edges, the situation analysis nodes execute corresponding calculation according to input situation data and output data, the decision nodes input the output data of the situation analysis nodes to make internal decisions and return to the next decision node needing to be decided, and the action nodes execute related actions according to decision results;
s2, performing simulation, judging the node type, if the node type is a decision node, performing corresponding decision and returning to the next decision node, if the next decision node is not empty, iterating, if the node type is an action node, directly performing corresponding action, and if the node type is a situation analysis node, performing topological sorting on all nodes, and performing each situation analysis node in sorting in a recycling manner to obtain a final output value;
the situation analysis node is provided with 0 or more output data edges, the action node is provided with and only has one output decision edge, the decision node is provided with and only has one input decision edge, the situation analysis node is provided with 0 or more output decision edges, the decision node receives the data output by the situation analysis node, and the decision node is provided with 0 or more input data edges;
the step of performing topology ordering on all the nodes in the step S2 specifically includes: initializing a sorting node list, acquiring all nodes with output, acquiring all nodes with input, calculating all node lists with output nodes but without input, calculating all node lists with input nodes but without output, returning to the sorting node list if all node lists with output nodes but without input are not empty, and outputting the sorted sorting node list;
the functional decision tree exists in the form of a Python script, and each element in the functional decision tree metamodel is mapped into a corresponding script element.
2. The functional equipment decision behavior simulation modeling system is characterized by comprising:
the construction module is used for constructing a functional decision tree, the functional decision tree comprises a meta-model root node, the meta-model root node comprises tree nodes and edges, the tree nodes comprise situation analysis nodes, decision nodes and action nodes, the edges comprise data edges for transmitting data and decision edges for decision control, the situation analysis nodes, the decision nodes and the action nodes are related through the data edges and the decision edges, the situation analysis nodes execute corresponding calculation according to input situation data and output data, the decision nodes input the output data of the situation analysis nodes to make internal decisions and return to the next decision node needing to be decided, and the action nodes execute related actions according to decision results;
the simulation execution module is used for executing simulation, judging the node type, executing corresponding decision and returning to the next decision node if the node type is the decision node, iterating if the next decision node is not empty, directly executing corresponding action if the node type is the action node, and performing topology sequencing on all nodes if the node type is the situation analysis node, and performing each situation analysis node in sequencing in a recycling way to obtain a final output value;
the situation analysis node is provided with 0 or more output data edges, the action node is provided with and only has one output decision edge, the decision node is provided with and only has one input decision edge, the situation analysis node is provided with 0 or more output decision edges, the decision node receives the data output by the situation analysis node, and the decision node is provided with 0 or more input data edges;
the step of topologically ordering all the nodes specifically includes: initializing a sorting node list, acquiring all nodes with output, acquiring all nodes with input, calculating all node lists with output nodes but without input, calculating all node lists with input nodes but without output, returning to the sorting node list if all node lists with output nodes but without input are not empty, and outputting the sorted sorting node list;
the functional decision tree exists in the form of a Python script, and each element in the functional decision tree metamodel is mapped into a corresponding script element.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210551852.4A CN114792072B (en) | 2022-05-18 | 2022-05-18 | Function-based equipment decision behavior simulation modeling method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210551852.4A CN114792072B (en) | 2022-05-18 | 2022-05-18 | Function-based equipment decision behavior simulation modeling method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114792072A CN114792072A (en) | 2022-07-26 |
| CN114792072B true CN114792072B (en) | 2024-01-16 |
Family
ID=82463435
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210551852.4A Active CN114792072B (en) | 2022-05-18 | 2022-05-18 | Function-based equipment decision behavior simulation modeling method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114792072B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117252335A (en) * | 2023-09-20 | 2023-12-19 | 杭州中微感联信息技术有限公司 | Machine learning-based municipal equipment facility intelligent management method and system |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110619131A (en) * | 2018-06-20 | 2019-12-27 | 陕西师范大学 | UUV agent behavior planning method based on weighted fuzzy Petri net |
| CN111476337A (en) * | 2020-04-02 | 2020-07-31 | 中国人民解放军国防科技大学 | Group robot mode generation and conversion method of multistage variable gene regulation and control network |
| CN112287521A (en) * | 2020-10-10 | 2021-01-29 | 东北大学 | Decision-making platform of intelligent combat equipment |
| CN112307622A (en) * | 2020-10-30 | 2021-02-02 | 中国兵器科学研究院 | Autonomous planning system and planning method for generating military forces by computer |
| CN112329348A (en) * | 2020-11-06 | 2021-02-05 | 东北大学 | Intelligent decision-making method for military countermeasure game under incomplete information condition |
| CN113392521A (en) * | 2021-06-11 | 2021-09-14 | 中国人民解放军军事科学院评估论证研究中心 | Method and system for constructing resource marshalling model for air-sea joint combat mission |
| CN113420462A (en) * | 2021-07-20 | 2021-09-21 | 西安九思智汇信息科技有限公司 | Intelligent target simulation system based on behavior tree |
| CN114115504A (en) * | 2022-01-27 | 2022-03-01 | 北京华清未来能源技术研究院有限公司 | Distributed power supply automatic alarm method and device based on Hadoop big data platform |
| CN114201885A (en) * | 2021-12-16 | 2022-03-18 | 航天物联网技术有限公司 | Improved behavior tree-based military force entity behavior simulation element modeling method and system |
| CN114373360A (en) * | 2021-12-17 | 2022-04-19 | 清华大学 | Intelligent training system, method and device for flight simulator |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2120192A1 (en) * | 2008-05-13 | 2009-11-18 | Sap Ag | Method and system for supporting a decision-making process |
| CA2834877A1 (en) * | 2012-11-28 | 2014-05-28 | Henry Leung | System and method for event monitoring and detection |
-
2022
- 2022-05-18 CN CN202210551852.4A patent/CN114792072B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110619131A (en) * | 2018-06-20 | 2019-12-27 | 陕西师范大学 | UUV agent behavior planning method based on weighted fuzzy Petri net |
| CN111476337A (en) * | 2020-04-02 | 2020-07-31 | 中国人民解放军国防科技大学 | Group robot mode generation and conversion method of multistage variable gene regulation and control network |
| CN112287521A (en) * | 2020-10-10 | 2021-01-29 | 东北大学 | Decision-making platform of intelligent combat equipment |
| CN112307622A (en) * | 2020-10-30 | 2021-02-02 | 中国兵器科学研究院 | Autonomous planning system and planning method for generating military forces by computer |
| CN112329348A (en) * | 2020-11-06 | 2021-02-05 | 东北大学 | Intelligent decision-making method for military countermeasure game under incomplete information condition |
| CN113392521A (en) * | 2021-06-11 | 2021-09-14 | 中国人民解放军军事科学院评估论证研究中心 | Method and system for constructing resource marshalling model for air-sea joint combat mission |
| CN113420462A (en) * | 2021-07-20 | 2021-09-21 | 西安九思智汇信息科技有限公司 | Intelligent target simulation system based on behavior tree |
| CN114201885A (en) * | 2021-12-16 | 2022-03-18 | 航天物联网技术有限公司 | Improved behavior tree-based military force entity behavior simulation element modeling method and system |
| CN114373360A (en) * | 2021-12-17 | 2022-04-19 | 清华大学 | Intelligent training system, method and device for flight simulator |
| CN114115504A (en) * | 2022-01-27 | 2022-03-01 | 北京华清未来能源技术研究院有限公司 | Distributed power supply automatic alarm method and device based on Hadoop big data platform |
Non-Patent Citations (4)
| Title |
|---|
| 基于情景分析的航空集群决策规则库构建方法;胡利平 等;《航空学报》;第41卷(第S1期);第37-52页 * |
| 基于攻击防御树的态势感知预警模型;毕锦雄;;《信息安全与通信保密》(第09期);第50-59页 * |
| 基于蒙特卡洛树搜索方法的空战机动决策;何旭 等;《火力与指挥控制》;第18卷(第03期);第36-41页 * |
| 面向装备效能仿真的战术探索方法研究;姚剑;《中国博士学位论文全文数据库 社会科学I辑》(第02期);第G112-1页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114792072A (en) | 2022-07-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hu et al. | Application of deep reinforcement learning in maneuver planning of beyond-visual-range air combat | |
| Virtanen et al. | Modeling air combat by a moving horizon influence diagram game | |
| CN112820164B (en) | VR virtual confrontation training system based on layered behavior model | |
| CN112329348A (en) | Intelligent decision-making method for military countermeasure game under incomplete information condition | |
| CN112560332B (en) | Aviation soldier system intelligent behavior modeling method based on global situation information | |
| CN113255916A (en) | Using method of combat knowledge system and combat decision auxiliary method | |
| CN113705102B (en) | Deduction simulation system, deduction simulation method, deduction simulation equipment and deduction simulation storage medium for sea-air cluster countermeasure | |
| CN112307622A (en) | Autonomous planning system and planning method for generating military forces by computer | |
| CN114792072B (en) | Function-based equipment decision behavior simulation modeling method and system | |
| CN110619131A (en) | UUV agent behavior planning method based on weighted fuzzy Petri net | |
| CN114169142A (en) | Task meta-model construction method, database and device for general combat process | |
| Smith Iii et al. | Autonomous and cooperative robotic behavior based on fuzzy logic and genetic programming | |
| Lundell et al. | Fuzzy petri net for uav decision making | |
| Deng et al. | Research on intelligent decision technology for multi-UAVs prevention and control | |
| Xiaochao et al. | A cgf behavior decision-making model based on fuzzy bdi framework | |
| Tyagi et al. | Application of Fuzzy Logic Control Systems in Military Platforms | |
| Tran et al. | Adaptation of a mamdani fuzzy inference system using neuro-genetic approach for tactical air combat decision support system | |
| Costa | The fighter aircraft's autodefense management problem: a dynamic decision network approach | |
| Wu et al. | Decision modeling and simulation of fighter air-to-ground combat based on reinforcement learning | |
| Mendi et al. | Applications of Reinforcement Learning and its Extension to Tactical Simulation Technologies | |
| Tao et al. | A function-based behavioral modeling method for air combat simulation | |
| Montgomery et al. | Abductive networks applied to electronic combat | |
| Pang et al. | Role-Based Bayesian Decision Framework for Autonomous Unmanned Systems | |
| CN114329928B (en) | Modularized assembly and overall parameter rapid generation method of equipment model | |
| Karelahti | Modeling and on-line solution of air combat optimization problems and games |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |