CN114117073B - Cluster intention recognition method and device, computer equipment and readable medium - Google Patents
Cluster intention recognition method and device, computer equipment and readable medium Download PDFInfo
- Publication number
- CN114117073B CN114117073B CN202111431123.7A CN202111431123A CN114117073B CN 114117073 B CN114117073 B CN 114117073B CN 202111431123 A CN202111431123 A CN 202111431123A CN 114117073 B CN114117073 B CN 114117073B
- Authority
- CN
- China
- Prior art keywords
- intention
- cluster
- entity
- fleet
- machine group
- 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 51
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 87
- 239000011159 matrix material Substances 0.000 claims description 25
- 238000012549 training Methods 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 9
- 238000013527 convolutional neural network Methods 0.000 claims description 9
- 230000000694 effects Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 abstract description 2
- 238000005755 formation reaction Methods 0.000 description 60
- 230000008569 process Effects 0.000 description 11
- 230000000875 corresponding effect Effects 0.000 description 8
- 230000008901 benefit Effects 0.000 description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 241000270295 Serpentes Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 229910003460 diamond Inorganic materials 0.000 description 1
- 239000010432 diamond Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Animal Behavior & Ethology (AREA)
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a method, a device, computer equipment and a readable medium for identifying intent of a cluster, wherein the method comprises the following steps: acquiring the position and model information of an airplane in a current fleet formation to be analyzed, and identifying the fleet pattern of the fleet formation according to the position information; acquiring a first fleet intent associated with a fleet pattern and a second fleet intent associated with each aircraft model from a fleet knowledge graph; exchanging the first machine group intention with the second machine group intention to obtain an initial intention recognition result; if the initial intention recognition result contains more than one machine group intention, calculating the real intention of the current machine group formation to be analyzed by adopting a Sagnac decision criterion; screening target entities associated with real intentions from the cluster knowledge graph and outputting the target entities; the application carries out the recognition of the intention of the machine group based on the knowledge graph, considers the influence of various factors to obtain the most probable intention of the machine group, improves the accuracy of the intention recognition and provides reference for command and control decision under the intelligent condition.
Description
Technical Field
The present application relates to the technical field of command control systems, and more particularly, to a method and apparatus for recognizing a fleet intention, a computer device, and a readable medium.
Background
With the rapid development of intelligent perception technology, the detection and recognition precision and accuracy of the aerial targets are greatly improved. Efficient, correct analysis of the air target intent can provide effective support for correct decisions, one of the key factors in gaining decision advantage. As the number of airborne targets increases rapidly and appears to be highly dynamic, it also presents challenges for proper identification of target intent. Because airborne targets are often organized in groups, commonly performing tasks, airborne clusters often have a relatively consistent intent.
In the aspect of group intention recognition, the existing research is mostly focused on recognizing formation forms through certain types of characteristics of the formation line forms, such as field characteristics and tactical group characteristics, as an important basis for recognizing target intention; there are few studies on how to perform intent analysis of targets based on identified fleet formations, and no means to incorporate "knowledge" has been incorporated to enhance the intent analysis effect.
Disclosure of Invention
Aiming at least one defect or improvement requirement of the prior art, the invention provides a method, a device, computer equipment and a readable medium for identifying the intention of a cluster, which are used for constructing a correlation model between the characteristic and the intention of the cluster on the basis of analyzing the characteristic and the intention of the cluster so as to establish a cluster knowledge map which can be expanded iteratively, and comprehensively identifying the intention of the cluster by combining a method of a Sagnac decision criterion in a decision theory so as to provide reference for command control under an intelligent condition.
To achieve the above object, according to a first aspect of the present invention, there is provided a method for recognizing a fleet intention, the method comprising the steps of:
acquiring the position and model information of an airplane in a current fleet formation to be analyzed, and identifying the fleet pattern of the fleet formation according to the position information;
acquiring a first fleet intention associated with the fleet pattern and a second fleet intention respectively associated with each aircraft model from a pre-established fleet knowledge map;
the first machine group intention is submitted to a second machine group intention to obtain an initial intention recognition result;
If the initial intention recognition result contains a cluster intention, taking the cluster intention as the real intention of the cluster formation to be analyzed currently;
If the initial intention recognition result contains more than one machine group intention, a savart decision criterion is adopted to screen out the real intention of the machine group formation to be analyzed currently from the multiple machine group intentions;
and searching a target entity associated with the real intention from the cluster knowledge graph and outputting the target entity.
Preferably, in the above method for recognizing a machine group intention, if any machine group intention does not exist in the initial intention recognition result, the first machine group intention and the second machine group intention are combined to obtain an updated initial intention recognition result.
Preferably, in the above method for identifying a fleet intention, the creating of the fleet knowledge graph includes:
respectively establishing an intention entity and a common entity, and defining the attribute of each entity; the intention entity is a group intention type entity; the common entity comprises an airplane type entity, a cluster formation array entity, an airplane model type entity, a departure place entity, a target type entity and a target entity;
and establishing association relations between the common entities and between the intention entities and the common entities.
Preferably, in the above method for identifying a fleet intention, the step of using savanii decision criteria to screen out real intention of a fleet formation to be analyzed from a plurality of fleet intentions includes:
Based on a savage decision criterion, constructing an intention discrimination income table of each cluster intention relative to the real intention in the initial intention recognition result;
establishing an intention judgment opportunity loss table based on the intention judgment income table to obtain an opportunity loss value of each cluster intention relative to the real intention in an initial intention recognition result;
searching the minimum maximum opportunity loss in the intention judging opportunity loss table, wherein the machine group intention corresponding to the minimum maximum opportunity loss is the real intention of the machine group formation.
Preferably, in the above method for recognizing a fleet intention, when the initial intention recognition result is presentIf the intention of the personal group is the intention to judge that the size of the profit table is/>The/>, of the tableLine/>Column element/>Represents the/>Group intent of column versus the/>A profit value for the true intent of the row; wherein n is a natural number greater than or equal to 2; i is more than or equal to 1 and less than or equal to n, and j is more than or equal to 1 and less than or equal to n.
Preferably, in the above method for identifying a fleet intent, the searching a target entity associated with the real intent from the fleet knowledge graph and outputting the target entity includes:
searching a target type set associated with the real intention from the cluster knowledge graph;
If any target type does not exist in the target type set, indicating that the cluster formation to be analyzed currently is not specific to the specific target type;
Otherwise, inquiring a target entity set associated with each target type in the target type set from the cluster knowledge graph, counting all target entities in the target entity set, which fall into the activity radius of the cluster formation, as target entities possibly aiming at the cluster formation to be analyzed currently, and outputting the target entities.
Preferably, in the above method for recognizing a fleet intent, the recognizing a fleet pattern of the fleet formation includes:
training the convolutional neural network by adopting a cluster formation matrix training set to obtain a trained cluster formation matrix structural feature model; the fleet formation matrix training set includes a plurality of training samples having different fleet formation matrix types;
and identifying the current cluster formation model to be analyzed according to the trained cluster formation model structure feature model.
According to a second aspect of the present invention, there is also provided a machine group intention recognition apparatus including:
the array type identification module is used for acquiring the position and model information of the aircraft in the current cluster formation to be analyzed and identifying the cluster array type of the cluster formation according to the position information;
The intention recognition module is used for acquiring a first fleet intention associated with the fleet type and a second fleet intention respectively associated with each aircraft model from a pre-established fleet knowledge graph; the first machine group intention is submitted to a second machine group intention to obtain an initial intention recognition result;
If the initial intention recognition result contains a cluster intention, taking the cluster intention as the real intention of the cluster formation to be analyzed currently;
If the initial intention recognition result contains more than one machine group intention, a savart decision criterion is adopted to screen out the real intention of the machine group formation to be analyzed currently from the multiple machine group intentions;
and the result output module is used for searching the target entity associated with the real intention from the cluster knowledge graph and outputting the target entity.
According to a second aspect of the present invention, there is also provided a computer device comprising at least one processing unit, and at least one storage unit, wherein the storage unit stores a computer program which, when executed by the processing unit, causes the processing unit to perform the steps of the computer program product of any one of the above.
According to a second aspect of the present invention there is also provided a computer readable medium storing a computer program executable by a computer device, the computer program, when run on the computer device, causing the computer device to perform the steps of the method for identifying a fleet intention as described in any of the preceding claims.
In general, the above technical solutions conceived by the present invention, compared with the prior art, enable the following beneficial effects to be obtained:
The method, the device, the computer equipment and the readable medium for identifying the fleet intention construct a fleet knowledge graph capable of reflecting the incidence relation of the fleet intention, the aircraft attribute, the departure place, the target and the fleet intention, better solve the incidence description among the fleet, the aircraft individuals and the intention and have iterative expansibility; all possible fleet intents are obtained from the fleet knowledge graph based on the identified fleet patterns of the fleet formation and all aircraft types contained in the fleet formation, and then the real intents are screened by utilizing a fleet intent comprehensive judgment method.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for identifying a fleet intent provided in the present embodiment;
FIG. 2 is a flow chart of the comprehensive recognition process of the fleet intent provided by the present embodiment;
FIG. 3 is a logic block diagram of a device for recognizing intent of a cluster according to the present embodiment;
Fig. 4 is a logic block diagram of a computer device according to the present embodiment.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application.
The terms first, second, third and the like in the description and in the claims and in the above drawings, are used for distinguishing between different objects and not necessarily for describing a particular sequential or chronological order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, article, or apparatus.
Because the aircraft are usually organized into clusters in a formation mode, the aircraft can cooperatively complete a unified task target, and therefore, the uniform group intention can be reflected. The knowledge graph has the capability of representing and analyzing complex relationships among entities, so that the attributes of the entities and the association relationships among the entities can be well described, and the knowledge graph has been widely applied to the fields of information recommendation, intelligent search, deep question-answering and the like. The invention provides a knowledge-based method for identifying the intention of a fleet, which mainly considers the intention represented by formation array type structural features of the fleet and the intention which can be represented by the functional attributes of an airplane, establishes the association between the features and the intention through a knowledge graph, and then comprehensively identifies the intention of the fleet. The specific technical scheme is as follows: training common matrix type structural features of the cluster by using a convolutional neural network algorithm to obtain a matrix type structural feature model for cluster formation matrix type recognition; performing association modeling on information such as the cluster array type, the attribute, the intention and the like to construct a cluster knowledge graph; querying a cluster knowledge graph through the identified cluster array type and airplane model information to obtain a possible intention set of the cluster; finally, the intent of the cluster is identified by a comprehensive judging method.
Fig. 1 is a flowchart of a method for identifying a fleet intent, which is shown in fig. 1, and mainly includes the following steps:
S1, acquiring the position and model information of the aircraft in the current fleet formation to be analyzed, and identifying the fleet pattern of the fleet formation according to the position information of each aircraft;
in the embodiment, the convolutional neural network based on deep learning is adopted to realize intelligent recognition of the cluster array; specific:
Training the convolutional neural network by adopting a cluster formation matrix training set to obtain a trained cluster formation matrix structural feature model; the fleet formation matrix training set comprises a plurality of training samples with different fleet formation matrix types; different fleet formation patterns include, but are not limited to, common patterns such as arrow patterns, diamond patterns, snake patterns, wedge patterns, longitudinal patterns, ladder patterns, transverse patterns and the like, and the training samples comprise images of various fleet formation patterns and movement directions thereof generated by airplane position information.
And identifying the current cluster formation model to be analyzed according to the trained cluster formation model structure feature model.
The training process of the cluster formation matrix structural feature model refers to the conventional training process of the convolutional neural network, and is not described in detail in this embodiment.
The cluster type recognition in the present embodiment may be performed by using the characteristics of the prior art based on the domain characteristics of the cluster queues and the task state attributes, and is not limited to the convolutional neural network.
S2, acquiring a first fleet intention associated with the fleet pattern and a second fleet intention respectively associated with each aircraft model from a pre-established fleet knowledge graph;
In this embodiment, intent analysis is performed based on a cluster knowledge graph which is created in advance and can be expanded iteratively, and the creation of the cluster knowledge graph includes:
Respectively establishing an intention entity and a common entity, and defining the attribute of each entity; the intention entity is a group intention type entity; the common entity comprises an airplane type entity, a cluster formation array entity, an airplane model type entity, a departure place entity, a target type entity and a target entity; and establishing association relations between the common entities and between the intention entities and the common entities.
Specifically, the detailed process of establishing the cluster knowledge association model is as follows:
Step 1.1, an airplane type entity is established and used for representing various airplane types such as fighters, conveyers, early warning planes, reconnaissance planes and the like, and the attribute of each airplane type is defined as a quantity representing the importance degree of the airplane type The value range is [0,1].
Step 1.2, a group type entity is established, and corresponds to the airplane type and is used for representing different types of airplane groups formed by different airplane types such as fighter plane groups, transport plane group scout plane groups and the like. Meanwhile, the addition of one type of mixed formation group represents the situation that different types of aircrafts are mixed and formed.
And 1.3, establishing a cluster formation matrix entity which is used for representing common cluster formation matrix types such as cross teams, wedge teams and the like, wherein the attribute of the cluster formation matrix entity is linked with a cluster formation matrix type structural feature model trained by a Convolutional Neural Network (CNN) method.
Step 1.4 builds an aircraft model type entity for representing a specific model of the aircraft, such as "model a", "model B", etc. For different types of aircrafts, the attribute of the aircrafts is related to various indexes in the development process of the aircrafts, including the maximum range, the movable radius, the maximum flying height, the purpose and the like; the usage attributes of each type of aircraft represent the tasks that it can perform, in relation to its intent.
Step 1.5 establishes a fleet intent type entity for representing different intents of combat, bombing, surveillance, early warning, airborne fueling, and the like. The attribute of the entity of the machine group intention type is defined as an amount reflecting the threat degree of the intention due to different threat degrees caused by different intentsThe value range is [0,1].
Step 1.6, a departure place entity is established and used for representing each airport, a water surface activity airport, a land maneuvering platform and the like which are departure places of the clusters, wherein the attribute of the departure place entity is the information of the departure places of the clusters, which is stored in a knowledge graph or updated in real time.
Step 1.7, establishing target type entities including ground targets, aerial targets, water surface targets and the like.
Step 1.8, establishing target entities for the my target entities such as various active water surface targets, air targets, airports, ports and the like, wherein the attributes comprise: the target position information comprises static target position longitude and latitude information stored by a knowledge graph or dynamic target current position information acquired and updated in real time.
Step 1.9, establishing various entity association relations, mainly comprising two types: the general entity relationship comprises the relationships of belongings, compositions, presentations, etc. among the model entities, the fleet formation matrix type entities, the target type entities and the target entities and between each entity and the attribute thereof; and the second is the association relation of the common entity to the intention entity, which comprises the support relation between the model entity and the intention entity, between each aircraft model entity and the intention entity, and the relation of the intention entity to the target type entity, such as fight intention to the air target, ground attack intention to the ground target and the like.
Then, comprehensively identifying the intent of the cluster based on the knowledge-graph associated query result, and specifically:
S21, the first machine group intention is submitted to the second machine group intention to obtain an initial intention recognition result;
Particularly, if any machine group intention does not exist in the initial intention recognition result, merging the first machine group intention with the second machine group intention to obtain an updated initial intention recognition result;
s22, if the initial intention recognition result contains a cluster intention, taking the cluster intention as the real intention of the cluster formation to be analyzed currently;
S23, if the initial intention recognition result comprises more than one machine group intention, screening out the real intention of the machine group formation to be analyzed currently from a plurality of machine group intentions by adopting a Sagnac decision criterion;
in this embodiment, a savart decision criterion is used to screen out real intention of a fleet formation to be analyzed from a plurality of fleet intentions, which specifically includes:
Based on a savage decision criterion, constructing an intention discrimination income table of each cluster intention relative to the real intention in the initial intention recognition result;
when the initial intention recognition result exists If the intention of the personal group is the intention to judge that the size of the profit table is/>The/>, of the tableLine/>Column element/>Represents the/>Group intent of column versus the/>A profit value for the true intent of the row; wherein n is a natural number greater than or equal to 2; i is more than or equal to 1 and less than or equal to n, and j is more than or equal to 1 and less than or equal to n.
Establishing an intention judgment opportunity loss table based on the intention judgment income table to obtain an opportunity loss value of each cluster intention relative to the real intention in an initial intention recognition result;
searching the minimum maximum opportunity loss in the intention judging opportunity loss table, wherein the machine group intention corresponding to the minimum maximum opportunity loss is the real intention of the machine group formation.
S4, searching a target entity associated with the real intention from the cluster knowledge graph and outputting the target entity; the method specifically comprises the following steps:
Searching a target type set associated with the real intention from the cluster knowledge graph;
If any target type does not exist in the target type set, indicating that the cluster formation to be analyzed currently is not specific to the specific target type;
otherwise, inquiring a target entity set associated with each target type in the target type set from the cluster knowledge graph, counting all target entities in the target entity set, which fall into the activity radius of the cluster formation, as target entities possibly aiming at the cluster formation to be analyzed currently, and outputting the target entities.
Fig. 2 is a schematic flow chart of a comprehensive recognition process of the intent of the fleet, which is provided in this embodiment, wherein the comprehensive recognition process of the intent of the fleet is respectively input and output as follows:
input: based on the knowledge graph correlation query result, acquiring an intention set correlated with the current cluster array type Intent/>, associated with various aircraft models included in a fleetThe fleet comprises a total number of aircraft types/>And of the type/>Aircraft type number/>(E.g., number of bombers of airplane type 2), origin information of the fleet.
And (3) outputting: judgment result of machine group intentionAnd target entities for which the cluster may be directed.
As shown in fig. 2, the specific steps of comprehensive identification are as follows:
step 2.1, intersection of the fleet intent associated with the fleet formation matrix entity obtained by query and the fleet intent associated with each aircraft model entity in the fleet is obtained, and an initial intent recognition result is obtained 。
Step 2.2, judging the initial intention recognition result,There may be three situations: case 1, ifIf there is only one intention of the cluster, the result of recognition/>Ending the comprehensive intention judgment process, and turning to step 2.7; case 2, if/>If there are a plurality of machine group intents, each machine group intention may be the final machine group intention, and go to step 2.4 to further judge; case 3, if/>I.e. there is no intersection of any cluster intents, all initial cluster intents may be the intent of the cluster, and go to step 2.3 for further processing.
Step 2.3, merging the fleet intent corresponding to the fleet formation matrix entity with the fleet intent corresponding to each aircraft model entity in the fleet, namely。
Step 2.4, based on the savart decision criteria, constructing an intention discrimination profit table of each cluster intention relative to the real intention in the initial intention recognition result.
If it isExist in/>Personal intention, then the revenue table size is/>The/>, of the tableLine/>Column element/>Represent the firstIntent estimate for column versus the/>A profit value for the true intent of the row; wherein n is a natural number greater than or equal to 2; i is more than or equal to 1 and less than or equal to n, and j is more than or equal to 1 and less than or equal to n.
When (when)When the benefit value is the/>The product of threat degree of true intention and the sum of the products of the number of various types of airplanes, namely/>, namely；
When (when)When the benefit value is the/>Line/>Column benefit value minus the/>Line/>Column benefit value, i.e./>。
And 2.5, establishing an intention judging opportunity loss table based on the intention judging income table, namely subtracting the element values of each row from the maximum value of each row in the intention judging income table to obtain the opportunity loss value of each cluster intention relative to the real intention in the initial intention recognition result.
Step 2.6, searching the minimum maximum opportunity loss to judge, counting the maximum value of each column of opportunity loss in the opportunity loss judging table, and searching the minimum of the opportunity loss in each column of the opportunity loss judging table, namely, searching the minimum maximum opportunity loss of the intention estimation result relative to the real intention, wherein the intention estimation result corresponding to the column is the final intention recognition result of the cluster, namely, the real intention.
Step 2.7, inquiring a cluster knowledge graph, and searching a target type set corresponding to the real intention as. If it isIndicating that the machine group intention is not specific to the specific target type, outputting a final intention recognition result, and turning to step 2.9; otherwise, the target entity set/>, corresponding to the associated query target type。
Step 2.8 statistics of target entity setsTarget entity/>, which falls within the cluster radius of activityThat is, all target entities/>, within the current cluster activity radius, can be calculated based on the cluster departure location information, the intersection of each model activity radius within the cluster (the activity radius is derived from the aircraft attribute), and the target entity location informationAs target information for which the current cluster intention recognition result may be aimed.
And 2.9, finishing the comprehensive identification process.
It should be noted that while in the above-described embodiments the operations of the methods of the embodiments of the present specification are described in a particular order, this does not require or imply that the operations must be performed in that particular order or that all of the illustrated operations be performed in order to achieve desirable results. Rather, the steps depicted in the flowcharts may change the order of execution. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step to perform, and/or one step decomposed into multiple steps to perform.
The embodiment also provides a device for identifying the intention of a cluster, as shown in fig. 3, which comprises a matrix type identification module, an intention identification module and a result output module, wherein,
The array recognition module is used for acquiring the position and model information of the aircraft in the current to-be-analyzed cluster formation and recognizing the cluster array type of the cluster formation according to the position information of each aircraft;
The intention recognition module is used for acquiring a first fleet intention associated with the fleet type and a second fleet intention respectively associated with each aircraft model from a pre-established fleet knowledge graph; the first machine group intention is submitted to a second machine group intention to obtain an initial intention recognition result;
If the initial intention recognition result contains a cluster intention, taking the cluster intention as the real intention of the cluster formation to be analyzed currently;
If the initial intention recognition result contains more than one machine group intention, a savart decision criterion is adopted to screen out the real intention of the machine group formation to be analyzed currently from the multiple machine group intentions;
And the result output module is used for searching the target entity associated with the real intention from the cluster knowledge graph and outputting the target entity.
For specific limitations on the means for recognizing the intent of the fleet, reference is made to the above limitations on the method for recognizing the intent of the fleet, and no further description is given here. The respective modules in the above-described fleet intent recognition device may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
The embodiment also provides a computer device, as shown in fig. 4, which includes at least one processor and at least one memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to execute the steps of the above-mentioned method for identifying the intent of the cluster, which is not repeated herein; in the present embodiment, the types of the processor and the memory are not particularly limited, for example: the processor may be a microprocessor, digital information processor, on-chip programmable logic system, or the like; the memory may be volatile memory, non-volatile memory, a combination thereof, or the like.
The computer device may also communicate with one or more external devices (e.g., keyboard, pointing terminal, display, etc.), with one or more terminals that enable a user to interact with the computer device, and/or with any terminals (e.g., network card, modem, etc.) that enable the computer device to communicate with one or more other computing terminals. Such communication may be through an input/output (I/O) interface. Moreover, the computer device may also communicate with one or more networks such as a local area network (Local Area Network, LAN), a wide area network (Wide Area Network, WAN), and/or a public network such as the internet via a network adapter.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SYNCHLINK) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
It will be readily appreciated by those skilled in the art that the foregoing description is merely a preferred embodiment of the invention and is not intended to limit the invention, but any modifications, equivalents, improvements or alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (8)
1. A method for identifying a fleet intent, comprising:
Acquiring the position information and the model information of the aircraft in the current cluster formation to be analyzed, and identifying the cluster matrix type of the cluster formation according to the position information;
acquiring a first fleet intention associated with the fleet pattern and a second fleet intention respectively associated with each aircraft model from a pre-established fleet knowledge map;
the first machine group intention is submitted to a second machine group intention to obtain an initial intention recognition result;
If the initial intention recognition result contains a cluster intention, taking the cluster intention as the real intention of the cluster formation to be analyzed currently;
if the initial intention recognition result contains more than one machine group intention, calculating the real intention of the current machine group formation to be analyzed by adopting a Sagnac decision criterion;
searching a target entity associated with the real intention from the cluster knowledge graph and outputting the target entity;
the creation of the cluster knowledge graph comprises the following steps:
respectively establishing an intention entity and a common entity, and defining the attribute of each entity; the intention entity is a group intention type entity; the common entity comprises an airplane type entity, a cluster formation array entity, an airplane model type entity, a departure place entity, a target type entity and a target entity;
establishing association relations between common entities and between intention entities and the common entities;
the method for calculating the real intention of the cluster formation to be analyzed currently by adopting the savart decision rule comprises the following steps:
Based on a savage decision criterion, constructing an intention discrimination income table of each cluster intention relative to the real intention in the initial intention recognition result;
establishing an intention judgment opportunity loss table based on the intention judgment income table, and calculating an opportunity loss value of each cluster intention relative to the real intention in an initial intention recognition result;
searching the minimum maximum opportunity loss in the intention judging opportunity loss table, wherein the machine group intention corresponding to the minimum maximum opportunity loss is the real intention of the machine group formation.
2. The method for recognizing a machine group intention according to claim 1,
If any machine group intention does not exist in the initial intention recognition result, merging the first machine group intention with the second machine group intention to obtain an updated initial intention recognition result.
3. The method for recognizing intent of cluster as recited in claim 1, wherein when the initial intent recognition result existsIf the intention of the personal group is the intention to judge that the size of the profit table is/>The/>, of the tableLine/>Column element/>Represents the/>Group intent of column versus the/>A profit value for the true intent of the row; wherein n is a natural number greater than or equal to 2; i is more than or equal to 1 and less than or equal to n, and j is more than or equal to 1 and less than or equal to n.
4. The method according to any one of claims 1 to 2, wherein the searching for a target entity associated with the real intention from the cluster knowledge graph and outputting the target entity includes:
searching a target type set associated with the real intention from the cluster knowledge graph;
If any target type does not exist in the target type set, indicating that the cluster formation to be analyzed currently is not specific to the specific target type;
Otherwise, inquiring a target entity set associated with each target type in the target type set from the cluster knowledge graph, counting all target entities in the target entity set, which fall into the activity radius of the cluster formation, as target entities possibly aiming at the cluster formation to be analyzed currently, and outputting the target entities.
5. The fleet intent recognition method as recited in any one of claims 1 to 2, wherein the recognizing the fleet pattern of the fleet formation includes:
training the convolutional neural network by adopting a cluster formation matrix training set to obtain a trained cluster formation matrix structural feature model; the fleet formation matrix training set includes a plurality of training samples having different fleet formation matrix types;
and identifying the current cluster formation model to be analyzed according to the trained cluster formation model structure feature model.
6. A fleet intent recognition device, comprising:
the array type identification module is used for acquiring the position information and the model information of the aircraft in the current cluster formation to be analyzed and identifying the cluster array type of the cluster formation according to the position information;
The intention recognition module is used for acquiring a first fleet intention associated with the fleet type and a second fleet intention respectively associated with each aircraft model from a pre-established fleet knowledge graph; the first machine group intention is submitted to a second machine group intention to obtain an initial intention recognition result;
If the initial intention recognition result contains a cluster intention, taking the cluster intention as the real intention of the cluster formation to be analyzed currently;
if the initial intention recognition result contains more than one machine group intention, calculating the real intention of the current machine group formation to be analyzed by adopting a Sagnac decision criterion;
The result output module is used for searching a target entity associated with the real intention from the cluster knowledge graph and outputting the target entity;
the creation of the cluster knowledge graph comprises the following steps:
respectively establishing an intention entity and a common entity, and defining the attribute of each entity; the intention entity is a group intention type entity; the common entity comprises an airplane type entity, a cluster formation array entity, an airplane model type entity, a departure place entity, a target type entity and a target entity;
establishing association relations between common entities and between intention entities and the common entities;
the method for calculating the real intention of the cluster formation to be analyzed currently by adopting the savart decision rule comprises the following steps:
Based on a savage decision criterion, constructing an intention discrimination income table of each cluster intention relative to the real intention in the initial intention recognition result;
establishing an intention judgment opportunity loss table based on the intention judgment income table, and calculating an opportunity loss value of each cluster intention relative to the real intention in an initial intention recognition result;
searching the minimum maximum opportunity loss in the intention judging opportunity loss table, wherein the machine group intention corresponding to the minimum maximum opportunity loss is the real intention of the machine group formation.
7. A computer device comprising at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program which, when executed by the processing unit, causes the processing unit to perform the steps of the machine group intention recognition method of any one of claims 1 to 5.
8. A computer readable medium, characterized in that it stores a computer program executable by a computer device, which when run on the computer device causes the computer device to perform the steps of the method for recognizing a machine intent of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111431123.7A CN114117073B (en) | 2021-11-29 | 2021-11-29 | Cluster intention recognition method and device, computer equipment and readable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111431123.7A CN114117073B (en) | 2021-11-29 | 2021-11-29 | Cluster intention recognition method and device, computer equipment and readable medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114117073A CN114117073A (en) | 2022-03-01 |
| CN114117073B true CN114117073B (en) | 2024-04-23 |
Family
ID=80371153
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111431123.7A Active CN114117073B (en) | 2021-11-29 | 2021-11-29 | Cluster intention recognition method and device, computer equipment and readable medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114117073B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115061501B (en) * | 2022-07-28 | 2022-11-22 | 中国人民解放军空军预警学院 | Method and system for identifying coded data of air formation |
| CN115238832B (en) * | 2022-09-22 | 2022-12-02 | 中国人民解放军空军预警学院 | CNN-LSTM-based air formation target intention identification method and system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111930793A (en) * | 2020-06-26 | 2020-11-13 | 西安电子科技大学 | Target behavior mining and retrieval analysis method, system, computer equipment and application |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11195057B2 (en) * | 2014-03-18 | 2021-12-07 | Z Advanced Computing, Inc. | System and method for extremely efficient image and pattern recognition and artificial intelligence platform |
-
2021
- 2021-11-29 CN CN202111431123.7A patent/CN114117073B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111930793A (en) * | 2020-06-26 | 2020-11-13 | 西安电子科技大学 | Target behavior mining and retrieval analysis method, system, computer equipment and application |
Non-Patent Citations (1)
| Title |
|---|
| 军用无人机飞控技术发展简述;蒋超，赵娜，闻子侠;中国军转民;40-42 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114117073A (en) | 2022-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10636169B2 (en) | Synthesizing training data for broad area geospatial object detection | |
| US10733759B2 (en) | Broad area geospatial object detection using autogenerated deep learning models | |
| Kaselimi et al. | A vision transformer model for convolution-free multilabel classification of satellite imagery in deforestation monitoring | |
| US20160379388A1 (en) | System and method for combining geographical and economic data extracted from satellite imagery for use in predictive modeling | |
| CN114117073B (en) | Cluster intention recognition method and device, computer equipment and readable medium | |
| US20190114333A1 (en) | System and method for species and object recognition | |
| Kass et al. | Improving area of occupancy estimates for parapatric species using distribution models and support vector machines | |
| Krysovatyy et al. | Economic Crime Detection Using Support Vector Machine Classification. | |
| CN114330509B (en) | Method for predicting activity rule of aerial target | |
| Brax | Anomaly detection in the surveillance domain | |
| Rivas-Perea et al. | Statistical and neural pattern recognition methods for dust aerosol detection | |
| Gokaraju et al. | Identification of spatio-temporal patterns in cyber security for detecting the signature identity of hacker | |
| Zhang et al. | ARIMA Model‐Based Fire Rescue Prediction | |
| Rubio et al. | RELand: Risk Estimation of Landmines via Interpretable Invariant Risk Minimization | |
| Chandrayan et al. | Atmospheric weather fluctuation prediction using machine learning | |
| Privalov et al. | Decision Support System for Integrated Geo-Information Environmental Monitoring | |
| Seymour | Capturing the full potential of the synthetic theater operations research model (STORM) | |
| Marreiros et al. | Progress in Artificial Intelligence: 21st EPIA Conference on Artificial Intelligence, EPIA 2022, Lisbon, Portugal, August 31–September 2, 2022, Proceedings | |
| Liu et al. | Automatically detect and track infrared small targets with kernel Fukunaga-Koontz transform and Kalman prediction | |
| Jharko et al. | On creating safety control systems for high operation risk plants | |
| Blacknell et al. | ATR of ground targets: Fundamentals and key challenges | |
| Andersson et al. | Towards crime rate prediction through street-level images and siamese convolutional neural networks | |
| Yan et al. | A data‐driven adaptive geospatial hotspot detection approach in smart cities | |
| Mindrup et al. | Extending robust parameter design to noise by noise interactions with an application to hyperspectral imagery | |
| Rai et al. | Criminal Activities Predictive Analysis Using Data Mining Techniques |
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 |