Data-driven complex system componentization modeling method and related product

Abstract

The scheme discloses a kind of complex system component modeling method based on data driving and related product, wherein the steps of the method include: complex equipment system component modeling and combat process constraint description;Combat process data and complex equipment system component modeling associated combat equipment system are analyzed by co-simulation, and the analysis results such as model checking, knowledge matching and conflict resolution of real combat process data and simulation data are obtained.The scheme can schedule the bottom equipment component model under the atomic behavior component model, thereby improving the behavior modeling efficiency and enhancing the reusability of atomic behavior component model.At the same time, based on the combat process constraint described by description logic, the model checking, knowledge matching and conflict resolution of real combat process data and simulation data are used to realize model confidence evaluation, constraint relationship analysis, rule matching analysis and model optimization.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of digital modeling, in particular to a data-driven complex system component modeling method and device, electronic equipment and readable storage medium. BACKGROUND

[0002] Foreign countries have actively researched data-driven complex system modeling methods since 2000. In 2007, a scholar proposed the "deep green" plan, in which the two important concepts of the "deep green" system are pre-estimation and adaptive execution. Pre-estimation connects the simulation system with the external command system, and in the process of combat, the simulation system is connected with the external combat process data to assist the combat process decision-making by capturing the combat process data. In 2010, a research laboratory conducted a "real-time action plan analysis using high-performance computing" project, and conducted data-driven simulation to support real-time decision-making systems. In recent years, this technology has been applied in various fields such as crisis management, engineering and manufacturing. Some domestic industry experts are also actively exploring and have published more than 20 articles such as "dynamic data-driven adaptive modeling and simulation". Some are also conducting research, for example, "production scheduling modeling and simulation based on multi-agent and dynamic data-driven", "dynamic data-driven highway pavement disease generation and evolution model research" and other topics. In terms of data-driven complex combat system modeling and simulation technology, there are currently few relevant reports and data in China. SUMMARY

[0003] The purpose of the present application is to provide a data-driven complex system component modeling method, device, electronic equipment and readable storage medium.

[0004] To achieve the above purpose, the technical scheme adopted by the present application is as follows:

[0005] In a first aspect, the present application provides a data-driven complex system component modeling method, and the steps of the method include:

[0006] The complex equipment system component model and the combat process constraint are described;

[0007] The combat equipment system associated with the combat process data and the complex equipment system component model is co-simulated and analyzed to obtain the analysis results of the model checking of the real combat process data and the simulation data, and / or the analysis results of the knowledge matching and conflict resolution.

[0008] In a second aspect, the present application provides a data-driven complex system component modeling device, and the device includes:

[0009] A combat process knowledge base is used to provide combat process constraints during modeling;

[0010] The complex equipment system componentization model is used for providing a behavior model in the modeling process;

[0011] The simulation analysis model block cooperatively simulates and analyzes the complex combat equipment system associated with the complex combat equipment system componentization model and the combat process data, obtains an analysis result of model checking of real combat process data and simulation data, and / or an analysis result of knowledge matching and conflict resolution.

[0012] In a third aspect, the present application provides a computer storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method described above.

[0013] In a fourth aspect, the present application provides a computing device, which comprises a processor and a memory for storing executable instructions of the processor.

[0014] The processor is configured to execute the method described above by executing the executable instructions.

[0015] The present application has the following advantages:

[0016] The present application is based on a pre-constructed combat process knowledge base, and the atomic behavior component model is used to schedule the bottom equipment component model to realize the smallest atomic behavior in logic, which cannot be divided. The atomic behavior component model is combined to construct a composite behavior component model, which can improve the behavior modeling efficiency and enhance the reusability of the atomic behavior component model.

[0017] The present application can be based on the combat process constraint knowledge base described by the logic, and the model checking, knowledge matching and conflict resolution of real combat process data and simulation data are used to realize the model confidence evaluation, constraint relationship analysis, rule matching analysis and model optimization. DETAILED DESCRIPTION

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort.

[0019] Figure 1 A schematic diagram showing an example of the data-driven complex combat system componentization modeling method described in the present application;

[0020] Figure 2 A schematic diagram showing an example of the complex equipment system componentization model framework described in the present application;

[0021] Figure 3A schematic diagram showing the combat behavior modeling process according to the present application is shown.

[0022] Figure 4 A schematic diagram showing the combat equipment system cooperative simulation analysis of data and model association according to the present application is shown.

[0023] Figure 5 An instance diagram showing the simulation formation mode based on formation rules according to the present application is shown.

[0024] Figure 6 A schematic diagram of the electronic device according to the present application is shown. DETAILED DESCRIPTION

[0025] In order to make the technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0026] Through analysis and research of the prior art, the data-driven modeling method technology is more advanced at present, and relevant units at home and abroad have explored the application research in related systems. However, with the characteristics of the next generation of combat system, such as the scale of equipment elements, flexible use of combat, complex interaction, strong game confrontation, etc., the existing modeling means are difficult to meet the needs, and it is necessary to explore the application of data-driven modeling method in complex system design and verification, break through the formal description of combat process data constraints, intelligent component behavior modeling method, process data and component model association modeling, and promote the application of data-driven modeling method in complex equipment system.

[0027] Therefore, the present scheme aims to provide a data-driven component modeling method, particularly a component modeling method for associating combat process data with complex equipment system models. To address the problems caused by complex causal relationships, lack of information, and dynamic updates in the modeling process of complex combat systems / systems, the present invention is based on big data, semantic networks, and artificial intelligence technologies. First, the combat process knowledge base is formed by describing the scene constraints, process constraints, performance constraints, and combat rule constraints of the equipment system combat process using description logic. On this basis, a combat-oriented component modeling method is proposed. The atomic behavior component model is used to schedule the underlying equipment component model, achieving the smallest atomic behavior in logic that cannot be divided. The atomic behavior component model can improve the efficiency of behavior modeling and enhance the reusability of atomic behavior component models by combining to construct composite behavior component models. Finally, based on the combat process constraint knowledge base described by description logic, the knowledge matching and conflict resolution methods of real combat process data and simulation data are used to realize constraint relationship analysis, model optimization, and system performance evaluation.

[0028] In the following, a data-driven component modeling method and device proposed by the present scheme will be described in detail in conjunction with the drawings. Specifically as follows:

[0029] 1. Combat process constraint description based on description logic

[0030] First, the elements composed of combat process data are determined and all mutual relationships between the elements are established. All constituent elements of the equipment meta-model of the present invention are equipment ontology, performance indicators, combat activities, action rules, equipment systems, system functions, system capabilities, and capability planning. Among them, the relationships between elements include: equipment entities and performance indicators, functional characteristics and performance indicators of equipment entities. The performance indicators of equipment entities determine their ability to complete tasks according to the role of functional characteristics in combat activities; equipment entities and combat activities, equipment entities are the basic units of combat activities. Combat activities consist of a series of actions performed by one or more equipment entities in a combat scene; equipment entities and equipment systems, a plurality of functionally independent equipment entities are connected to form an equipment system; equipment entities and combat rules, equipment entities need to follow combat rules in the combat process; performance indicators and system indicators, system indicators are a series of performance indicators combined according to functional performance indicators; equipment systems and system functions, the capabilities of equipment systems are reflected in system functions; equipment systems and combat activities, equipment systems participate in combat activities. Combat activities consist of a series of actions performed by one or more equipment systems in a combat scene; capability planning and system capability, the expected capability of the system is expressed by capability planning, and the actual system capability is measured by capability planning; combat activities and combat rules, combat activities are executed according to the requirements of combat rules.

[0031] 2. Componentized model of complex equipment system

[0032] The componentized model framework of complex equipment system is shown in Figure 2 It mainly includes: combat process characteristic database, equipment component model library, bottom equipment component model, intelligent combat behavior model, public service function, simulation engine. The intelligent combat behavior model is divided into atomic behavior model, composite behavior model and cognitive behavior model.

[0033] Combat process characteristic database: classified according to two dimensions of data content and management level, classified according to data content into my information data, enemy information data, white side data, battlefield environment data and comprehensive data, classified according to management level into combat basic data, real-time situation data and decision support data. The above data will be sent to each component model, including intelligent behavior component model and bottom equipment component model, through the way of scenario parameter binding or guided control in the simulation process.

[0034] Bottom equipment component model library: a single bottom equipment component model is the basic unit of the equipment system. In actual combat, different equipment performs different tasks, and the equipment types in the bottom equipment model library are divided into reconnaissance, attack, control and communication.

[0035] Equipment component model: a single bottom equipment component model can further construct a higher level equipment component model. For example, the aircraft formation model, which is composed of different warplanes such as early warning aircraft and fighter aircraft.

[0036] Intelligent combat behavior model: the intelligent combat behavior model as shown in Figure 3 includes both composite behavior model, which combines atomic behavior based on composite specification, and cognitive behavior model, which realizes flexible behavior decision mechanism. The cognitive behavior model adopts knowledge-based behavior decision model, which can be represented as a rule set:

[0037] RuleSet = <ID, Meta, Ins, Outs, Consts, Funs, Rules>

[0038] Wherein, ID is the identification of the combat rule set; Meta is the meta information of the rule; Ins is the input parameter set of the rule set; Outs is the output parameter set of the rule set; Consts is the defined constant symbol set; Funs is the operation function to get the current combat action; Rules is the member rule.

[0039] Through detection and analysis of historical data and current real-time data, the target is learned and inferred, and each parameter adaptive adjustment identification system based on the monitoring result is developed. On the basis of effective, reliable and robust perception of the target, cognition, feedback, adjustment strategy and decision-making are quickly completed. Intelligent combat behavior models are constructed in multiple dimensions such as time, space, frequency and polarization, thereby greatly improving the decision-making performance of the system. The specific method is as follows:

[0040] 1) The comprehensive integration method of multi-view classifier is adopted to combine different types of data at different levels. Through the cascade integration of classifiers (such as random forest, decision tree, SVM, sparse regression, etc.) in multiple dimensions and feature perspectives, target recognition is realized, thereby supporting subsequent combat decision-making;

[0041] 2) Deep reinforcement learning algorithm is studied to automatically generate a model for the optimal decision under the current state. Currently, there are two types of reinforcement learning methods. The first type is to solve the policy function, i.e. to learn the mapping function from state to action. This policy function can be modeled using machine learning methods such as deep learning. The second type is to solve the value function, i.e. to learn the expected return of taking action in a state under a certain policy. The goal of value function learning is to maximize the return value. Reinforcement learning based on value function learning is suitable for cases where the number of actions is limited. However, when the number of actions cannot be exhausted, most work adopts the method of policy-based learning. In combat decision-making based on the current situation, both the situation and the behavior are difficult to represent in a limited space, so the policy-based reinforcement learning method is adopted for intelligent decision-making.

[0042] The methods of combat process target recognition based on multiple dimension classifiers and combat decision generation based on reinforcement learning use asynchronous distributed parallel training architecture to perform parallel simulation training on large-scale multi-instances, realize parallel simulation and parallel model training for combat, and improve the intelligence level of situation awareness, target recognition and combat decision-making of the command and control system.

[0043] Public service functions: mainly include statistical learning models such as support vector machines and neural networks. For subsystems whose mechanism is unknown or completely unknown, the capability boundary of the subsystem is analyzed, and it is used as a constraint condition. Combined with a large amount of historical data information in the whole life cycle of equipment, statistical modeling method is used to start from the historical data of the whole life cycle of equipment, according to the number of sample data, neural network and support vector machine two machine learning methods are selected for data-driven equipment model modeling.

[0044] Simulation engine: in the running preparation stage, it is responsible for decomposing the combat behavior model until the atomic level behavior model; creating atomic behavior model instances; creating equipment model instances, i.e. simulation objects; in the running stage, the simulation objects are uniformly scheduled by the scheduler of the simulation engine, the input of the simulation object comes from the external simulator, the guiding system, the actual system or the monitoring system, and the simulation model credibility is evaluated online during the simulation running process. The simulation results can be serialized to the resource library by the simulation engine, or output to the visualization system. After the running is completed, the interaction data of the agent can be recorded by the engine throughout the process, and the simulation evaluation tool loads the evaluation model to analyze and evaluate the simulation results.

[0045] 3. Data and model associated combat equipment system cooperative simulation analysis

[0046] Firstly, based on the combat process meta-model, the equipment system and intelligent combat behavior modeling and simulation are implemented, then the model verification is implemented based on simulation data and real data to realize the confidence evaluation of the model, finally the conflict resolution of knowledge matching of real combat rules and combat process behavior is realized based on the description logic of combat process constraint description, realizing constraint relationship analysis, rule matching analysis and model optimization.

[0047] Firstly, based on the combat process meta-model, the equipment model involved in the simulation is created, then the behavior model is created, then the simulation running time parameters, the underlying equipment model are configured, and finally the simulation running is realized, and the simulation data is generated. As shown in Figure 2 The scheme further provides a modeling and simulation device implemented in cooperation with the above modeling method, as a complex equipment system componentized model framework. The device comprises a combat process characteristic database, an equipment component model library, an underlying equipment component model, an intelligent combat behavior model, a public service function and a simulation engine. The intelligent combat behavior model is divided into an atomic behavior model, a composite behavior model and a cognitive behavior model.

[0048] Then, the simulation data and the real data are checked to realize the model credibility evaluation. The model checking is divided into static checking and dynamic checking. The static checking data (also called sample) is the data produced by sampling, and its main feature is that a set of data meets the independent and identically distributed condition and has nothing to do with time, and the order of the data can be exchanged. The dynamic checking data is a set of observation values arranged in time sequence, and here it refers to the discrete sequence obtained after sampling at equal intervals. The static checking method of the simulation data and the real data includes the mean and variance comparison and the probability distribution comparison. When the mean and variance of the simulation data and the real data meet the index requirements, the two sets of data are considered to be consistent; when the probability distribution of the simulation data and the real data meets the index requirements, the two sets of data are considered to be consistent. The dynamic checking method of the simulation data and the real data includes the characteristic point comparison method and the grey correlation method. The characteristic point comparison method compares the average distance between the characteristic points of the two spatial curves of the simulation data and the real data to compare the two curves; the grey correlation method analyzes the shape closeness of the curves of the two sets of data to compare the data sequence as a whole.

[0049] Finally, the simulation combat behavior is matched with the combat process knowledge based on the rule constraint to realize the rule matching degree analysis, the constraint relationship analysis and the model optimization. The logical programming tool Racer is used as the conflict resolution tool in the scheme. The logical programming is a programming paradigm, which is mainly based on formal logic. It is designed for symbolic computation, and is particularly suitable for solving problems involving objects and their relationships. The Racer logical program is composed of a set of logical facts and rules. For example, the fact that "Adam is the father of Bob" can be written in the logical program as: father(adam, bob). Here, the parent class is the name of an attribute, called a predicate; Adam and Bob are its parameters. The rule that if the father of B is A, then A is also the relative of B can be written as: parent(A, B) :- father(A, B). Here, ":-" represents logical implication, and A and B are logical variables. By using selective linear definition (SLD) clause resolution, it can perform first-order logic reasoning. For example, taking the above fact and rule as a logical program, the following question can be asked to the Racer system:?- parent(X, bob). Racer can access the facts asserted in the previous program, and it will answer: X = adam. However, if the query is:?- parent(eve, bob), the answer will be: false, because the previous program does not specify any relationship between eve and bob. Due to its understandability and ability to perform first-order logic reasoning, it is used for knowledge matching, conflict resolution, etc.

[0050] Racer can handle some complex rule matching, constraint analysis problems, such as formation rule matching. Figure 5The instance object represents the matching of formation rules based on real formation rules, real formation patterns, and simulation formation patterns using the Racer tool. (a) represents the instance object. (b) represents five possible row / column formation rules. In a complex task setting, each formation rule is a separate combat rule. (c) is a simple simulation formation pattern matching rule with three real formation patterns. If the image contains three objects, and the outer two objects have the same shape and direction, the "between" relationship is true. If there is an object in the last row of the three objects that has the same shape and direction as the (single) object in the first row, the "occurs" relationship is true. If the image contains two objects with the same shape and direction, the "same" relationship is true. (d) is an example of two complex tasks, row formation rule matching (left) and column formation rule matching (right). The purpose of the row formation matching task (left) is to test whether two rows of objects have the same pattern according to the five formation combat rules in (b). The purpose of the column matching task (right) is to determine whether the formation pattern id of each example (2, 4, 3) corresponds to the corresponding pattern id in (b). The task of each example is to confirm whether the column of objects meets the specified five formation pattern ids. The combination of multiple shapes and relationships makes the variation of this task quite diverse. It should be understood that the modules or units in the scheme can be realized by hardware, software, firmware, or a combination thereof. In the above embodiments, a plurality of steps or methods can be realized by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application specific integrated circuits with suitable combination logic gate circuits, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

[0051] On the basis of the above-mentioned modeling method embodiments, the scheme further provides a computer-readable storage medium. The computer-readable storage medium is used to implement the program product of the above-mentioned modeling method, which can adopt a portable compact disc read-only memory (CD-ROM) and include program codes, and can be run on a device, such as a personal computer. However, the program product of the scheme is not limited thereto, and in this document, the readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, device, or apparatus.

[0052] The program product can employ any combination of one or more computer-readable media. The computer-readable media can be a computer-readable storage medium or a computer-readable signal medium. The computer-readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium include the following: an electrical connection having one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0053] The computer-readable storage medium can include a data signal embodied in a carrier wave, or propagated by a carrier wave. Such a propagated signal can take any of a variety of forms, including but not limited to electro-magnetic, optical, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium that is not a storage medium or is not a transmission medium.

[0054] Program code embodied on a computer-readable storage medium can be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0055] Program code for carrying out operations of the present scheme can be written in any combination of one or more programming languages, including an object-oriented programming language such as Java, C++, etc., or conventional procedural programming languages, such as the "C" programming language, or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device can be connected to the user computing device by any kind of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, by connecting to an Internet service provider via the Internet).

[0056] Based on the above-mentioned modeling method embodiments, the present scheme further provides an electronic device. As shown in Figure 6 The electronic device shown is merely an example and should not impose any limitation on the functions and use range of the embodiments of the present scheme.

[0057] As shown in Figure 6As shown, the electronic device 201 is in the form of a general computing device. Components of the electronic device 201 can include, but are not limited to, at least one storage unit 202, at least one processing unit 203, a display unit 204, and a bus 205 for connecting the different system components.

[0058] The storage unit 202 stores program code that can be executed by the processing unit 203 to cause the processing unit 203 to perform the steps of the various exemplary embodiments described in the device symptom information acquisition method. For example, the processing unit 203 can perform the steps of the modeling described above.

[0059] The storage unit 202 can include volatile storage units, such as random access memory (RAM), and / or non-volatile storage units, such as read-only memory (ROM).

[0060] The storage unit 202 can further include a program / utility, having a program module such as an operating system, one or more application programs, other program modules, and program data, each of which can implement aspects of a network environment, for example, as some or all of the exemplary embodiments.

[0061] The bus 205 can include a data bus, an address bus, and a control bus.

[0062] The electronic device 201 can also communicate with one or more external devices 207, such as a keyboard or a pointing device, through an input / output (I / O) interface 206. It will be appreciated that other hardware and / or software modules can be used in conjunction with the electronic device 201, such as microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.

[0063] It is apparent that the above-described embodiments of the present application are only examples for clearly and completely describing the present application and are not intended to limit the embodiments of the present application. It will be obvious to those skilled in the art that various changes and modifications can be made to the embodiments described above without departing from the spirit and scope of the present application. Therefore, the scope of the present application should not be limited to the embodiments described above but should be defined by the claims.

Claims

1. A data-driven based complex system componentized modeling method, characterized in that, The method comprises the following steps: Modeling of a complex equipment system componentization model and description of a combat process constraint; Co-simulation analysis of a combat equipment system associated with combat process data and the complex equipment system componentization model, to obtain analysis results of model checking of real combat process data and simulation data, and analysis results of knowledge matching and conflict resolution; The complex equipment system componentization model comprises the following steps: Building a bottom-layer equipment component model and an intelligent combat behavior model; The bottom-layer equipment component model comprises a plurality of basic units constituting an equipment system; The intelligent combat behavior model comprises a combination behavior model, a cognitive behavior model, and an atomic behavior model; The combat process constraint comprises: A semantic-based equipment system combat process meta-model; At least one constraint condition of equipment system combat is described based on description logic, a combat process knowledge base is formed, and the combat process knowledge base is taken as a constraint of the combat process; The at least one constraint condition of equipment system combat described based on description logic comprises scene constraint, flow constraint, performance constraint, and combat rule constraint of the equipment system combat process described based on description logic; The co-simulation analysis of the combat equipment system associated with the combat process data and the complex equipment system componentization model, to obtain the analysis results of model checking of real combat process data and simulation data, and the analysis results of knowledge matching and conflict resolution, comprises the following steps: Description of the combat process constraint based on description logic; Based on the pre-set model checking, logic reasoning tool, and simulation engine, model checking analysis results of the equipment and behavior model, and analysis results of knowledge matching and conflict resolution are obtained. The cognitive behavior model is a behavior decision-making model using knowledge, and the rule set thereof is:

2. The complex system componentized modeling method of claim 1, wherein, RuleSet=<ID, Meta, Ins, Outs, Consts, Funs, Rules> ID is an identifier of the combat rule set; Meta is meta-information of the rule set; Ins is a set of input parameters of the rule set; Outs is a set of output parameters of the rule set; Consts is a set of defined constant symbols; Funs is an operation function for obtaining a current combat action; and Rules is a member rule. The device comprises:

3. A data-driven based complex system componentized modeling apparatus, characterized in that, A combat process knowledge base for providing a combat process constraint in the modeling process, wherein the combat process constraint comprises a semantic-based equipment system combat process meta-model, at least one constraint condition of equipment system combat is described based on description logic, a combat process knowledge base is formed, and the combat process knowledge base is taken as a constraint of the combat process; and the at least one constraint condition of equipment system combat described based on description logic comprises scene constraint, flow constraint, performance constraint, and combat rule constraint of the equipment system combat process described based on description logic. ​ The complex equipment system component model is used for providing a behavior model in modeling, and a description of the complex equipment system component model comprises: constructing a bottom equipment component model and an intelligent combat behavior model; the bottom equipment component model comprises a plurality of basic units constituting an equipment system; the intelligent combat behavior model comprises a combination behavior model, a cognitive behavior model and an atomic behavior model; The simulation analysis model block is used for co-simulation analysis of a combat equipment system associated with complex equipment system component model and combat process data, to obtain an analysis result of model checking of real combat process data and simulation data and an analysis result of knowledge matching and conflict resolution, comprising: combat process constraint description based on description logic; based on a pre-set model checking, logic reasoning tool and simulation engine, a model checking analysis result of equipment and behavior model and an analysis result of knowledge matching and conflict resolution are obtained.

4. The complex system componentized modeling apparatus according to claim 3, wherein, The cognitive behavior model is a behavior decision model using knowledge, and a rule set thereof is: RuleSet=<ID, Meta, Ins, Outs, Consts, Funs, Rules> wherein ID is an identification of a combat rule set; Meta is meta information of the rule set; Ins is an input parameter set of the rule set; Outs is an output parameter set of the rule set; Consts is a defined constant symbol set; Funs is an operation function for obtaining a current combat action; and Rules is a member rule.

5. A computer storage medium, characterized in that, A computer program is stored thereon, and the program is executed by a processor to implement the method of claim 1 or 2.

6. A computing device, comprising: Comprise: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the method of claim 1 or 2 by executing the executable instructions.

Images