A hybrid granularity event-based naval battle reasoning method

Abstract

The application discloses a sea battle reasoning method based on mixed granularity events, which comprises the following steps: determining important terms and concept words in the field and application range of ontology according to knowledge data of a battle knowledge base; establishing a vocabulary list according to the important terms and concept words; defining the ontology class and the hierarchy of the class, determining the parent-child relationship and parallel relationship between the classes according to the vocabulary list; defining the attributes and attribute constraints of the class, and obtaining a sea battlefield combat event ontology model; implementing instantiation on the sea battlefield mixed granularity combat information through the sea battlefield combat event ontology model, obtaining a plurality of sea battlefield combat events and storing the events in a battle rule knowledge base; reasoning the obtained sea battlefield combat events according to a BN reasoning engine module, and obtaining a sea battlefield combat event to be executed at the next moment. The method is convenient for describing the key events, semantic elements and the relationship between the events of the sea battlefield combat, and greatly improves the efficiency of the deduction of the sea battlefield task through the analysis of the military field knowledge and the decision-making knowledge of the commander.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sea battlefield situation data, and particularly relates to a sea battle reasoning method based on mixed granularity events. BACKGROUND

[0002] Task planning and action plan deduction evaluation is a deduction of a series of sequenced actions that a combat platform subject needs to complete at a certain time and place under certain conditions, which describes an action deduction method of forming a system dynamics-oriented model through the formalization of military field knowledge analysis and commander decision-making knowledge.

[0003] Modern battlefield technology is intensive, combat forces have shifted from single-service operations to joint operations, and weapons and equipment are surrounded by current advanced knowledge and innovative technologies, so combat forces are complex and changeable, such as reconnaissance capabilities, reconnaissance and surveillance ranges of high-tech reconnaissance equipment can cover the entire battlefield, for example, attack capabilities, the range (range) of various weapons and equipment has increased, the range of artillery is up to 70 kilometers, the range of tactical missiles is up to 2000 kilometers, and the combat radius of aircraft is up to 1800 kilometers. In the face of such multi-dimensional combat fields and complex combat logic, the traditional action deduction method has a variety of event representation models, including MUC event templates, ACE event structures, ontology-based event models, and ABC model methods, which are difficult to meet the needs of battlefield evolution. That is, in the context of modern information warfare, the amount of information that commanders need to process when conducting combat operations has increased dramatically, not only do they need to quickly analyze and judge massive amounts of data, but they also need to weigh real-time information resources and quickly deduce and evaluate them before issuing battlefield decisions. Such situations make it more difficult for traditional methods to cope with rapid changes in the battlefield. SUMMARY

[0004] The present application provides a sea battle reasoning method based on mixed granularity events to overcome the above technical problems.

[0005] To achieve the above-mentioned purpose, the technical solution of the present application is:

[0006] A sea battle reasoning method based on mixed granularity events, comprising the following steps:

[0007] Step S1: determining the field and application range of the ontology according to the knowledge data of the war knowledge base, wherein the war knowledge base comprises a weapon platform knowledge base and a battle rule knowledge base;

[0008] Step S2: obtaining important terms and concept words in the field ontology according to the field and application range of the ontology, and establishing a vocabulary list according to the important terms and concept words;

[0009] Step S3: defining a plurality of ontology classes and class hierarchy according to the vocabulary list, determining parent-child relationship and parallel relationship between classes; and defining attributes and attribute constraints of the classes;

[0010] Step S4: establishing a sea battlefield combat event model of the ontology classes; defining a plurality of entity atomic behaviors and entity atomic states of the ontology classes, and setting sea battlefield hybrid granularity combat information based on the entity atomic behaviors and the entity atomic states;

[0011] Step S5: instantiating the sea battlefield hybrid granularity combat information through the sea battlefield combat event ontology model, obtaining a plurality of sea battlefield combat events and storing them in a battle rule knowledge base;

[0012] Step S6: reasoning the obtained sea battlefield combat events according to the BN reasoning engine module, obtaining a sea battlefield combat event to be executed at the next time of each sea battlefield combat event.

[0013] Further, the plurality of ontology classes defined according to the vocabulary list include action classes, event coding classes, time classes, location classes, state classes, and participant classes;

[0014] The action classes include a behavior sequence of an entity in a sea battlefield event within a certain time period;

[0015] The event coding classes include numbered subclasses of each sea battlefield combat event;

[0016] The time classes include a starting time point subclass of an event, a time zone subclass of an event, and a duration subclass of an event;

[0017] The location classes include a spatial position subclass of a sea battlefield event;

[0018] The state classes include a current sea battlefield event subclass and other sea battlefield event subclasses associated with the current sea battlefield event subclass;

[0019] The participant classes include a fixed-wing aircraft subclass, a ground base subclass, a rotary-wing aircraft subclass, an underwater vessel subclass, and a surface vessel subclass.

[0020] Further, the setting of the sea battlefield hybrid granularity combat information in step S4 is

[0021] Step S4.1: determining a plurality of ontology classes participating in a sea battlefield combat event, and setting entity atomic behaviors and entity atomic states of each of the ontology classes;

[0022] The entity atomic behaviors are minimum granularity actions completed by combat entities, and each entity atomic behavior is serially or in parallel constructed into an action behavior sequence;

[0023] The entity atom state is a state condition of the combat entity itself, and the state condition includes but is not limited to the speed, the ammunition carrying amount and the duration of the combat entity itself;

[0024] And a plurality of event ontology information are established based on the entity atom behavior and the entity atom state of each ontology class;

[0025] Step S4.2: An event relationship sequence list is established based on the time and space features of each event ontology information;

[0026] Step S4.3: Attribute constraints are performed on the attributes of the event ontology information ontology class in the event relationship sequence list, and naval battlefield mixed granularity combat information is obtained

[0027] Further, the attributes of the defined class include object attributes and data attributes;

[0028] And the object attributes represent the attributes of the relationship between classes, including

[0029] The Precede represents before; the Follow represents after; the HasAction represents the owned action; the HasState represents the owned state; the HasPlace represents the place where the event occurs; and the HasCode represents the owned identification;

[0030] The numerical attribute represents the relationship between instances or between an instance and a numerical value, including

[0031] String represents that the instance content corresponding to the class is a text, and the numerical type corresponding to the class is a character type;

[0032] Integer represents that the instance content corresponding to the class is a numerical value or data, and the numerical type needs to be limited to an integer;

[0033] Float represents that the instance content corresponding to the class is a numerical value or data, and the numerical type needs to be limited to a floating point;

[0034] Time represents that the instance content corresponding to the class is a date or a time, and needs to be corresponding to a time;

[0035] The attribute constraints include has attribute constraints and is a attribute constraints;

[0036] The has attribute constraint is an extended attribute representing possession; the is a attribute constraint is used between two classes; and one of the classes represents a parent class, and the other class represents a child class.

[0037] Further, the naval battlefield combat event model is:

[0038] ,

[0039] wherein, representing the dynamic event that occurs; representing the participants of the event, divided into initiators and bearers; representing the time of the event; P represents the place of the event; H represents how the event occurs; E represents the encoding of the event.

[0040] Further, the BN reasoning engine module includes a pattern matcher, an agenda unit, and an execution engine unit;

[0041] The pattern matcher is used to match the sea battlefield combat events stored in the battle law rule knowledge base; and the matched sea battlefield combat events are transmitted to the agenda unit;

[0042] The agenda unit determines the priority of the sea battlefield combat events based on the matched sea battlefield combat events; and the sea battlefield combat events with determined priority are transmitted to the execution engine unit;

[0043] The execution engine unit executes the sea battlefield combat events in order based on the determined priority.

[0044] Beneficial effects: the present application provides a sea battle reasoning method based on mixed granularity events, including the following steps: determining the field and application range of the ontology according to the knowledge data of the battle knowledge base, obtaining important terms and concept words in the field ontology according to the field and application range of the ontology, and establishing a vocabulary list according to the important terms and concept words; defining the ontology class and the hierarchy of the class, determining the parent-child relationship and parallel relationship between the classes, and defining the attributes and attribute constraints of the class, obtaining the sea battlefield combat event ontology model; through the sea battlefield combat event ontology model, the sea battlefield mixed granularity combat information is instantiated, a plurality of sea battlefield combat events are obtained and stored in the battle law rule knowledge base; according to the BN reasoning engine module, the obtained sea battlefield combat events are reasoned, and the next time execution of the sea battlefield combat events is obtained. The key events, semantic elements and relationships between events of the sea battlefield combat are described, and the efficiency of the analysis of the military field knowledge and the reasoning of the sea battlefield task by the commander's decision-making knowledge is greatly improved. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0046] Figure 1 is a flow chart of the sea battle reasoning method of the present application;

[0047] Figure 2 An exemplary diagram of mixed granularity battle information acquisition in a sea battlefield in an embodiment of the present application;

[0048] Figure 3 An exemplary diagram of a hierarchy of a top class of a sea battlefield combat ontology in an embodiment of the present application;

[0049] Figure 4 An exemplary diagram of a hierarchy of a time class of a sea battlefield combat ontology in an embodiment of the present application;

[0050] Figure 5 An exemplary diagram of a class hierarchy of a sea battlefield combat ontology participant in an embodiment of the present application;

[0051] Figure 6 An exemplary diagram of a class hierarchy of a sea battlefield combat information ontology in an embodiment of the present application;

[0052] Figure 7 An exemplary diagram of object properties of a sea battlefield combat information ontology in an embodiment of the present application;

[0053] Figure 8 Data properties of a sea battlefield combat information ontology in an embodiment of the present application;

[0054] Figure 9 An exemplary diagram of a sea battlefield combat ontology model in an embodiment of the present application. DETAILED DESCRIPTION

[0055] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in connection with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0056] The embodiment provides a sea battle reasoning method based on mixed granularity events, as shown in the following steps: Figure 1

[0057] Step S1: determining the field and application range of the ontology according to knowledge data of a battle knowledge base, wherein the battle knowledge base comprises a weapon platform knowledge base and a battle rule knowledge base;

[0058] ​The battle knowledge base provides a knowledge data source for reasoning, wherein the weapon platform entity knowledge base stores combat unit entity information in a hierarchical structure, including equipment entities and equipment components, and equipment entity models are organized through a class-derived inheritance relationship in an organizational structure, which is mainly used to define performance parameters and self characteristics of the equipment entities and equipment components, thereby serving as a basis for parameterized configuration, and the equipment entities and components are managed according to a set hierarchy and organizational relationship, and the hierarchy structure and system classification are controlled by an xml file, and different equipment classes are defined according to different equipment entities and components. The performance parameters and self characteristics of the equipment components include an external size of the weapon, a weight of the weapon, a launch mode of the weapon, a launch preparation time of the weapon, an attack distance of the weapon, a hitting accuracy of the weapon, a guidance capability of the weapon, a survivability of the weapon, a penetration capability of the weapon, a service life of the weapon, and transportation and maintenance performance of the weapon.

[0059] The battle method rule knowledge base stores standardized battle rules for combat decision-making, and assists in reasoning decision-making. The battle method rule knowledge base includes a battle method model library and a real-time database. The battle method model library stores classic battle method templates in the battle field, and is used to assist the reasoning machine in decision support. The real-time database stores real-time battle simulation or perceived dynamic and static data of battle entities. In the construction of the battle method model library, behavior rules are constructed, and combat states, conditions and events are decomposed and refined through a hierarchical decomposition mechanism, and are specifically decomposed to specific atomic actions required by the events and evolution of the event state after the conditions are met. The state set of the entity model and the conversion conditions between the states are combed, and the state evolution rules are formally described, thereby providing knowledge support for the hybrid reasoning of the reasoning engine.

[0060] Step S2: According to the field and application range of the ontology, important terms and concept words in the field ontology are acquired, and a vocabulary list is established according to the important terms and concept words.

[0061] Step S3: According to the vocabulary list, a plurality of ontology classes and a class hierarchy are defined, parent-child relationships and parallel relationships between the classes are determined, and attributes and attribute constraints of the classes are defined.

[0062] Step S4: A naval battle field combat event model of the ontology class is established, entity atomic behaviors and entity atomic states of a plurality of ontology classes are defined, and hybrid granularity battle information of the naval battle field is set based on the entity atomic behaviors and the entity atomic states.

[0063] The entity atomic behavior is a minimum granularity action completed by a combat entity, and each entity atomic behavior is serially or in parallel constructed into an action behavior sequence.

[0064] The entity atomic state is a state condition of the combat entity itself, and the state condition includes but is not limited to the speed, the ammunition carrying amount and the duration of the combat entity itself;

[0065] The domain expert decomposes the task of the ontology class in the naval battlefield combat event model by the behavior editing component in a top-down manner, each step of decomposition decomposes the task of the ontology class into a plurality of subtasks in the form of a flowchart, a finite state machine or a rule set, and continues to decompose the subtasks until each subtask only completes a basic action of an entity, and the decomposition work is completed; the subtask as a leaf node is described by an atomic behavior model, the atomic behavior model obtained by decomposition includes an entity atomic behavior model and an entity atomic behavior state model; and the atomic behavior model is implemented by a programmer in an executable dynamic link library; the subtask of a non-leaf node and the task of a top layer node are described by a combined behavior model; a certain combined behavior model is assembled by the atomic behavior model and / or the combined behavior model decomposed in a lower layer; the atomic behavior model is expressed by an XML format behavior model description file, and basic data of each atomic behavior model is set; each atomic behavior model includes an equipment platform category, a formation level category and a provided capability equipment category; each atomic behavior model established by decomposition is saved into a model library, the file architecture of each atomic behavior model is implemented by a programmer, each atomic behavior model file is serially or in parallel constituted into an action behavior sequence file, and is saved to the model library for standby;

[0066] Step S5: the naval battlefield mixed granularity combat information is instantiated by the naval battlefield combat event ontology model, a plurality of naval battlefield combat events are obtained and stored in a battle rule knowledge base;

[0067] Step S6: each naval battlefield combat event is reasoned according to a BN reasoning engine module, and a naval battlefield combat event executed at a next time of each naval battlefield combat event is obtained.

[0068] According to the analysis of the characteristics of the naval battlefield combat information based on the knowledge data of the battle knowledge base, the classes of the naval battlefield combat information ontology and the hierarchical structure of the classes are determined. Through the analysis of the classes of the naval battlefield combat ontology and the relationship between the classes, the object attributes and the data attributes of the ontology model are determined. After the classes and the hierarchical structure of the classes and the related attributes of the ontology model are determined, the naval battlefield combat information ontology model is established by Protégé, and the ontology model is instantiated in combination with the related domain knowledge. The key events, semantic elements and relationships between events of the naval battlefield combat are facilitated to be described, and the efficiency of the analysis of the naval battlefield task by the military domain knowledge and the decision-making knowledge of the commander is greatly improved.

[0069] In specific embodiments, as Figure 3As shown, the plurality of ontology classes defined according to the vocabulary list includes an action class, an event code class, a time class, a location class, a state class, and a participant class;

[0070] The sea battlefield operation information contains operation events and time and space descriptions of the operation events and descriptions of participating entities in the sea battlefield. In Protégé, the sea battlefield operation information is defined by being divided into six classes, and the defined ontology classes include an action class, an event code class, a time class, a location class, a state class, and a participant class;

[0071] The action class includes a behavior sequence of an entity in a sea battlefield event within a time period;

[0072] The action class describes the behavior of an entity in a sea battlefield event within a time period. An action has a certain duration in time, and an action can also be refined into a sequence in granularity.

[0073] The event code class includes numbered subclasses of various sea battlefield operation events;

[0074] The event code class describes the numbering of a sea battlefield operation event, which is a unique number identifying a sea battlefield operation event and has uniqueness. An event code can uniquely identify each sea battlefield operation event;

[0075] The time class includes a start time point subclass of an event, a time zone subclass of an event, and a duration subclass of an event;

[0076] Time describes the time attribute of a sea battlefield event. The time class consists of three subclasses, namely a start time point, a time zone, and a duration. The specific hierarchical structure consists of Figure 4 As shown. The start time point of an event is composed of year, month, day, hour, minute, and second. The time zone describes the time zone of the location where the sea battlefield event occurs. The duration describes the duration of the entire process of the sea battlefield event from start to end;

[0077] The location class includes a spatial position subclass of a sea battlefield event;

[0078] The location class describes the spatial attribute of a sea battlefield event, which is a spatial description of a sea battlefield event and describes the location where the sea battlefield event occurs.

[0079] The state class includes a current sea battlefield event subclass and other sea battlefield event subclasses associated with the current sea battlefield event subclass;

[0080] The state class describes the behavior of an entity in a sea battlefield event at a certain moment, and the state is instantaneous in time, mainly showing whether the entity determines the occurrence of a certain behavior at a certain moment. The state usually follows a certain behavior, confirms whether the result of the behavior is successful, and then a new behavior will follow the state;

[0081] The participant class includes a fixed-wing aircraft subclass, a ground base subclass, a rotorcraft subclass, a submarine subclass, and a surface ship subclass;

[0082] The participant class describes the entities participating in sea battlefield operations. The participating entities are divided into two categories: the initiator of the event in the combat event and the recipient of the event in the combat event. The hierarchical structure of the recipient and the initiator is shown in Figure 5 The participant and the recipient are composed of four categories: fixed-wing aircraft, ground base, rotorcraft, submarine, and surface ship. Fixed-wing aircraft are divided into four categories: reconnaissance aircraft, fighter aircraft, unmanned reconnaissance aircraft, and air refueling aircraft. Rotorcraft are divided into two categories: reconnaissance helicopters and armed helicopters. Submarines are currently the only category of underwater vessels. Surface ships are divided into six categories: patrol ships, frigates, aircraft carriers, supply ships, light cruisers, and destroyers. Ground bases are divided into ten categories: reserve centers, arms depots, power plants, command centers, airports, weather stations, sea radar stations, logistics warehouses, power stations, and radar stations.

[0083] In specific embodiments, the properties of the defined class include object properties and data properties;

[0084] And the object properties represent the properties of the relationship between classes, including Precede, Follow, HasAction, HasState, HasPlace, and HasCode. Protégé defines and describes object properties using owl:ObjectPropery, rdfs:domains describes the definition domain of the property, and rdfs:Ranges describes the value domain of the property;

[0085] For the classes of the sea battlefield operation information ontology, the class and the property relationship between the classes designed in this paper mainly include Precede (before), Follow (after), HasAction (owned action), HasState (owned state), HasPlace (occurrence place), and HasCode (owned identification code), as shown in Table 1;

[0086] Table 1. Main object properties of sea battlefield operation information ontology

[0087]

[0088] The numerical attribute represents the relationship between instances or between instances and numerical values, including String, Integer, Float, and Time;

[0089] The numerical attribute describes a certain characteristic of a single class, that is, the attribute can be used to describe the relationship between instances or between instances and numerical values, for example, the attribute is called a data attribute if it limits the instance to a specific value or data format. Generally speaking, if the instance content corresponding to the class is a literal, the numerical type is a character type (string), if the instance content corresponding to the class is a numerical value or data, the numerical type needs to be limited to an integer (integer) or a floating point (float), and if the instance content corresponding to the class is a date or time, it needs to be corresponding to a time (Time), as shown in Table 2.

[0090] Table 2. Main data attributes of sea battlefield operation information ontology

[0091]

[0092] In specific embodiments, the setting of the sea battlefield hybrid granularity operation information in step S4 is

[0093] Step S4.1: Determine a plurality of ontology classes participating in the sea battle operation event, and set the entity atomic behavior and entity atomic state of each ontology class;

[0094] and establish a plurality of event ontology information based on the entity atomic behavior and entity atomic state of each ontology class;

[0095] Step S4.2: Establish an event relationship sequence list based on the time and space characteristics of each event ontology information; around the elements in the sea battlefield operation event, when designing the sea battlefield operation event ontology, both the classes and attributes of the sea battlefield operation event ontology should be able to accurately express the elements of the event, and the event ontology model should also be able to provide rich semantic operation information. Therefore, the sea battlefield operation event ontology around the event information and event relationship content includes; event information: in order to represent the event, the objects participating in the event need to be described first. Since the event occurs in a specific time and space, the time element also contains the time and space location characteristics of the action, and the main event information defined by the sea battlefield operation event ontology is shown in Table 3:

[0096] Table 3. Event information table of sea battlefield operation event ontology

[0097]

[0098] Event relationship: events and actions contained therein can be related to other events or actions. These relationships can be temporal and spatial coexistence, or have a time sequence, etc.

[0099] A series of relational descriptions enable the model to have the ability to describe, associate, and organize events in a fine-grained manner. The main event relationships defined in the naval battlefield combat event ontology are shown in Table 4.

[0100] Table 4. Event Relationship Table for Naval Battlefield Operations

[0101]

[0102] Step S4.3: Apply attribute constraints to the attributes of the event ontology information ontology class in the event relationship sequence list to obtain mixed-granularity combat information of the naval battlefield.

[0103] Once the relationship between event information and event information is determined, such as Figure 2 As shown, to ensure the completeness of the event ontology model, attribute constraints need to be applied to event information. In naval battlefield combat events, two attributes, `has` and `is a`, are defined. The `has` attribute is actually an extended attribute representing "ownership". For example, if `Event` is a naval battlefield combat event class, a constraint attribute `hasAtTime` needs to be defined, meaning the combat event must occur at a certain time. The `is a` attribute defines the concept of a lower-level class; if two classes have an `is a` attribute, then one class is the parent class and the other is the child class. By defining classes, attributes, and constraints between classes, mixed-granularity combat information from the naval battlefield can be obtained.

[0104] In a specific embodiment, the naval battlefield combat event ontology model is as follows:

[0105] ,

[0106] in, Indicates a dynamic event that has occurred; The participants in an event are categorized as initiators and sufferers. P indicates the time when the event occurred; H indicates the location when the event occurred; H indicates how the event occurred, mainly describing the actions of the event; E indicates the code of the event.

[0107] Specifically, the hybrid-granularity combat information of the maritime battlefield is implemented through the aforementioned maritime battlefield combat event ontology model, specifically as follows:

[0108] After clarifying the overall structure of the naval battlefield combat information ontology, the Protégé tool was used to construct the ontology. Based on the description of the naval battlefield combat information ontology above, the class of the naval battlefield combat information ontology constructed using Protégé is as follows: Figure 6 As shown:

[0109] After the classes of the sea battlefield operation information ontology are built in Protégé, the classes and the relationships between the classes are built in Object Properties, as shown in Figure 7 . .

[0110] After the classes and the relationships between the classes of the sea battlefield operation information ontology are built in Protégé, the data properties are built in Data Properties, as shown in Figure 8 . .

[0111] After the classes, object properties and data properties of the sea battlefield operation event ontology are built in Protégé, the construction of the sea battlefield operation ontology is preliminarily completed, and the sea battlefield operation ontology is as shown in Figure 9 . .

[0112] By combining the domain knowledge of the sea battlefield operation, the event elements of the sea battlefield operation event are determined; OWL DL is used as the modeling language of the sea battlefield operation information ontology, Protégé 5.5.0 is used as the tool for ontology modeling, and the seven-step method proposed by Stanford University is used as the basic process for constructing the sea battlefield operation information ontology. By combining the related domain knowledge to analyze the characteristics of the sea battlefield operation information, the classes of the sea battlefield operation information ontology and the hierarchical structure of the classes are determined. By analyzing the classes of the sea battlefield operation ontology and the relationships between the classes, the object properties and the data properties of the ontology model are determined. After the classes and the hierarchical structure of the classes and the related properties of the ontology model are determined, Protégé is used for modeling. After the sea battlefield operation information ontology model is built in Protégé, the ontology model is instantiated in combination with the related domain knowledge; .

[0113] In specific embodiments, the BN reasoning engine module is used to reason the acquired sea battlefield operation events, and the BN reasoning engine module includes a pattern matcher, an agenda unit and an execution engine unit; .

[0114] The pattern matcher is used to match the sea battlefield operation events stored in the battle method rule knowledge base, and the matched sea battlefield operation events are transmitted to the agenda unit; .

[0115] The agenda unit determines the priority of the sea battlefield operation events based on the matched sea battlefield operation events, and the sea battlefield operation events with the determined priority are transmitted to the execution engine unit; .

[0116] The execution engine unit executes the multiple sea battle combat events according to the determined priority order. Specifically, the rule engine (BN inference engine module) can acquire the event objects (facts) and the sea battle combat rules corresponding to the event objects in the battle rule template, acquire the confidence value of each sea battle combat rule by using the probabilistic soft logic algorithm, match the event objects satisfying the preset confidence threshold and the corresponding sea battle combat rules by using the pattern matcher (Pattern Matcher), and put the matched results into the agenda unit (Agenda). The agenda unit Agenda manages the matched sea battle combat rules. If multiple sea battle combat rules exist and execution conflicts occur, the agenda unit Agenda manages the execution order of the conflict rules according to the given priority order. Finally, the execution engine executes the rule activity part (RHS) of the rules in the agenda unit Agenda. The execution engine unit executes the rules according to the priority order in the agenda.

[0117] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A hybrid-granularity event-based naval battle reasoning method, characterized in that, The method comprises the following steps: Step S1: determining the field and application range of the ontology according to the knowledge data of the battle knowledge base, wherein the battle knowledge base comprises a weapon platform knowledge base and a battle rule knowledge base; Step S2: acquiring important terms and concept words in the field ontology according to the field and application range of the ontology, and establishing a vocabulary list according to the important terms and concept words; Step S3: defining a plurality of ontology classes and class hierarchies according to the vocabulary list, determining the parent-child relationship and parallel relationship between the classes, and defining the attributes and attribute constraints of the classes; The plurality of ontology classes defined according to the vocabulary list comprise an action class, an event coding class, a time class, a location class, a state class and a participant class; The action class comprises a behavior sequence of an entity in a time period in a sea battlefield event; The event coding class comprises a numbered subclass of each sea battlefield combat event; The time class comprises a starting time point subclass of an event, a time zone subclass of an event and a duration subclass of an event; The location class comprises a spatial position subclass of a sea battlefield event; The state class comprises a current sea battlefield event subclass and other sea battlefield event subclasses associated with the current sea battlefield event subclass; The participant class comprises a fixed-wing aircraft subclass, a ground base subclass, a rotary-wing aircraft subclass, an underwater vessel subclass and a surface vessel subclass; Step S4: establishing a sea battlefield combat event model of the ontology class, defining entity atomic behaviors and entity atomic states of the plurality of ontology classes, and setting sea battlefield hybrid granularity combat information based on the entity atomic behaviors and the entity atomic states; Step S5: implementing the sea battlefield hybrid granularity combat information through the sea battlefield combat event ontology model, acquiring a plurality of sea battlefield combat events and storing the plurality of sea battlefield combat events in the battle rule knowledge base; Step S6: reasoning each sea battlefield combat event according to a BN reasoning engine module, and acquiring a sea battlefield combat event to be executed at the next time of each sea battlefield combat event.

2. The hybrid-granularity event-based naval battle reasoning method according to claim 1, wherein, The setting of the sea battlefield hybrid granularity combat information in step S4 is specifically Step S4.1: determining a plurality of ontology classes participating in a sea battlefield combat event, and setting entity atomic behaviors and entity atomic states of each of the ontology classes; The entity atomic behavior is a minimum granularity action completed by a combat entity, and each entity atomic behavior is serially or in parallel connected to form an action behavior sequence; The entity atomic state is a state condition of the combat entity itself, and the state condition comprises but is not limited to the speed, ammunition carrying amount and duration of the combat entity itself; and a plurality of event ontology information are established based on the entity atomic behaviors and the entity atomic states of each ontology class; Step S4.2: establishing an event relationship sequence list based on the time and space characteristics of each event ontology information; Step S4.3: performing attribute constraints on the attributes of the event ontology information ontology classes in the event relationship sequence list, and obtaining sea battlefield hybrid granularity combat information.

3. The hybrid-granularity event-based naval battle reasoning method according to claim 1, wherein, The definition of the attributes of the classes comprises object attributes and data attributes; and the object attributes represent the attributes of the relationship between classes, comprising The Precede represents before; Follow represents after; HasAction represents the action of having; HasState represents the state of having; HasPlace represents the place of occurrence; HasCode represents the identification of having; The numerical attribute represents the relationship between instances or between instances and numerical values, including String represents the instance content of the class corresponding to the literal type as the character type; Integer indicates that the instance content of the class corresponds to a numerical value or data, and the numerical value type needs to be limited to an integer; Float indicates that the instance content of the class corresponds to a numerical value or data, and the numerical value type needs to be limited to a floating point; Time indicates that the instance content of the class corresponds to a date or time, and needs to correspond to a time; The attribute constraint includes has attribute constraint and is a attribute constraint; The has attribute constraint is an extended attribute indicating possession; The is a attribute constraint is used between two classes; And one of the classes represents the parent class; The other class represents the child class.

4. The hybrid-granularity event-based naval battle reasoning method according to claim 1, wherein, The sea battlefield combat event model is: wherein, represents a dynamic event that occurs; represents a participant in the event, divided into initiators and bearers; represents the time at which the event occurs; P represents the place at which the event occurs; H represents how the event occurs; and E represents the encoding of the event.

5. The hybrid-granularity event-based naval battle reasoning method according to claim 1, wherein, The BN reasoning engine module includes a pattern matcher, a agenda unit and an execution engine unit; The pattern matcher is used to match the sea battlefield combat events stored in the battle rule knowledge base; And the matched sea battlefield combat events are transmitted to the agenda unit; The agenda unit determines the priority of the sea battlefield combat events based on the matched sea battlefield combat events; And the sea battlefield combat events with determined priority are transmitted to the execution engine unit; The execution engine unit executes the sea battlefield combat events in order based on the determined priority.

Images