CN111125448A - Large-scale aerial task decision method and system - Google Patents
Large-scale aerial task decision method and system Download PDFInfo
- Publication number
- CN111125448A CN111125448A CN201911337772.3A CN201911337772A CN111125448A CN 111125448 A CN111125448 A CN 111125448A CN 201911337772 A CN201911337772 A CN 201911337772A CN 111125448 A CN111125448 A CN 111125448A
- Authority
- CN
- China
- Prior art keywords
- full
- decision rule
- area decision
- decision
- rule
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9014—Indexing; Data structures therefor; Storage structures hash tables
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The application provides a large-scale aerial task decision method, which comprises the following steps: acquiring a known air task decision knowledge system; constructing a non-full-area decision rule of an air task decision according to a known air task decision knowledge system; generating a full-area decision rule according to the non-full-area decision rule and filling the full-area decision rule into a Hash space; and executing the decision of the air task according to the whole area decision rule in the hash space. Compared with the prior art, the method and the device have the advantages that the human knowledge in the prior art is used as the basis, the human knowledge is analyzed into the plurality of incompletely-divided decision rules, then the completely-divided permutation and combination decision rules are automatically generated and automatically mapped to the Hash solution space, the problem of rule filling of a decision system is solved, and an efficient solution is provided for high-dimensional large-scale airplane air task decision.
Description
Technical Field
The application belongs to the technical field of airplane control, and particularly relates to a large-scale aerial task decision method and a large-scale aerial task decision system.
Background
The hash table, as a data structure for fast querying data, has a query time of O (1), and therefore is widely applied In a strong real-time decision system, because of the characteristics of the hash table, i.e. each decision point needs to list the result of the hash rule, when filling In the hash rule according to human knowledge, a combinatorial explosion problem is generated, i.e. if the dimension of the decision input vector is too high and the input area of each dimension is too many, according to the multiplication principle, a huge solution space growth occurs, e.g. the dimension of the input vector is n-dimension, the area of each dimension of the input vector is divided into I1, I220≈106And cannot completely fill human knowledge into solution space.
Disclosure of Invention
It is an object of the present application to provide a large scale over the air task decision method and system to solve or mitigate at least one of the problems of the background art.
In one aspect, the technical solution provided by the present application is: a large-scale over-the-air task decision-making method, the method comprising:
acquiring a known air task decision knowledge system;
constructing a non-full-area decision rule of an air task decision according to a known air task decision knowledge system;
generating a full-area decision rule according to the non-full-area decision rule and filling the full-area decision rule into a Hash space;
and executing the decision of the air task according to the whole area decision rule in the hash space.
In this application, the variable dimension of the full-region decision rule is greater than or equal to the variable dimension of the non-full-region decision rule.
In a preferred embodiment of the present invention, when the variable dimension of the full-area decision rule is greater than the variable dimension of the non-full-area decision rule, in the process of generating the full-area decision rule according to the non-full-area decision rule, the full-area decision rule is constructed with a parameter corresponding to a variable dimension which is the same as the variable dimension of the non-full-area decision rule and the variable dimension which is less than the variable dimension of the full-area decision rule, and the parameter corresponding to the variable dimension which is the same as the variable dimension of the non-full-area decision rule is ignored.
In a preferred embodiment of the present invention, when the variable dimension of the full-area decision rule is equal to the variable dimension of the non-full-area decision rule, in the process of generating the full-area decision rule according to the non-full-area decision rule, the full-area decision rule is constructed by using a parameter corresponding to the variable dimension of the non-full-area decision rule which is the same as the variable dimension of the full-area decision rule.
In a preferred embodiment of the present application, the hash space is a hash table.
In another aspect, the present application provides a large-scale over-the-air mission decision system, the system comprising:
the knowledge acquisition module is used for acquiring a known air task decision knowledge system;
the system comprises a non-whole area decision rule generating module, a whole area decision rule generating module and a whole area decision rule generating module, wherein the non-whole area decision rule generating module is used for constructing a non-whole area decision rule of an aerial task decision according to a known aerial task decision knowledge system;
a whole area decision rule generating module, configured to generate a whole area decision rule according to the non-whole area decision rule and fill the whole area decision rule into a hash space;
and the decision module is used for executing the decision of the air task according to the whole area decision rule in the Hash space.
In this application, the variable dimension of the full-region decision rule is greater than or equal to the variable dimension of the non-full-region decision rule.
In a preferred embodiment of the present invention, when the variable dimension of the full-area decision rule is greater than the variable dimension of the non-full-area decision rule, in the process of generating the full-area decision rule according to the non-full-area decision rule, the full-area decision rule is constructed with a parameter corresponding to a variable dimension which is the same as the variable dimension of the non-full-area decision rule and the variable dimension which is less than the variable dimension of the full-area decision rule, and the parameter corresponding to the variable dimension which is the same as the variable dimension of the non-full-area decision rule is ignored.
In a preferred embodiment of the present invention, when the variable dimension of the full-area decision rule is equal to the variable dimension of the non-full-area decision rule, in the process of generating the full-area decision rule according to the non-full-area decision rule, the full-area decision rule is constructed by using a parameter corresponding to the variable dimension of the non-full-area decision rule which is the same as the variable dimension of the full-area decision rule.
In a preferred embodiment of the present application, the hash space is a hash table.
Compared with the prior art, the method and the device have the advantages that the human knowledge in the prior art is used as the basis, the human knowledge is analyzed into the plurality of incompletely-divided decision rules, then the completely-divided permutation and combination decision rules are automatically generated and automatically mapped to the Hash solution space, the problem of rule filling of a decision system is solved, and an efficient solution is provided for high-dimensional large-scale airplane air task decision.
Drawings
In order to more clearly illustrate the technical solutions provided by the present application, the following briefly introduces the accompanying drawings. It is to be expressly understood that the drawings described below are only illustrative of some embodiments of the invention.
Fig. 1 is a schematic diagram of a large-scale air task decision method according to the present application.
Fig. 2 is a task decision diagram according to an embodiment of the present application.
Fig. 3 is a schematic diagram of a large-scale air mission decision system according to the present application.
Detailed Description
In order to make the implementation objects, technical solutions and advantages of the present application clearer, the technical solutions in the embodiments of the present application will be described in more detail below with reference to the drawings in the embodiments of the present application.
The hash table (also called hash table) is a data structure directly accessed according to key values, a given table M has a function f (key), and if an address recorded in the table containing a key can be obtained after a function is substituted into any given key, the given table M is called a hash table, and the function f (key) is a hash function, and the key is mainly characterized in that the access complexity is O (1).
In the field of aircraft executing an aerial mission decision, because decisions are real-time, it is necessary to ensure that computation performed on each decision is as small as possible, and multiple stages of logical decision branches will result in a longer decision time, so a hash table is usually used to perform lookup of O (1) time, that is, a mathematical abstraction of a decision system is a mapping of S- > D, where S is a state set and D is a decision set, and each time a decision system receives an environment input, that is, a state vector X ═ X1, X2, X3 … xn, it is necessary to output a corresponding decision vector Y ═ Y1, Y2, Y3, … … ym }, and a method for solving the mapping problem using the hash table is to uniquely encode the state vector X, for example, X ═ height, speed ═ speed, and whether there is an angle advantage }, and then the encoding is: and (4) taking the code as a key value, inputting a hash table and searching a corresponding decision mapping.
However, although the existing decision system based on the hash table can output the decision value corresponding to the code quickly, the following problems exist: 1) when the input state dimension is high and the number of variables corresponding to each dimension is large, the mapping relation combination of the hash table explodes, for example, a 20-dimensional air combat problem decision unit is solved, and after decision tree structure dependence modeling is carried out, 7 ten thousand hash mapping rules are generally needed; 2) it is very difficult to map the human knowledge into the hash table, and the corresponding decision outputs of the codes need to be filled one by one.
To this end, as shown in fig. 1, the present application proposes a large-scale air task decision method, including:
s1, acquiring a known air task decision knowledge system;
s2, constructing a non-full-area decision rule of the air task decision according to a known air task decision knowledge system;
s3, generating a full-region decision rule according to the non-full-region decision rule and filling the full-region decision rule into a Hash space;
and S4, executing the decision of the over-the-air task according to the full-region decision rule in the hash space.
In the present application, the variable dimension of the full-region decision rule is greater than or equal to the variable dimension of the non-full-region decision rule.
In the present application, when the variable dimension of the full-region decision rule is greater than the variable dimension of the non-full-region decision rule, in the process of generating the full-region decision rule according to the non-full-region decision rule, the full-region decision rule is constructed with parameters corresponding to the variable dimension which is the same as the variable dimension of the non-full-region decision rule and the full-region decision rule, and the parameters corresponding to the variable dimension which is smaller than the variable dimension of the full-region decision rule of the non-full-region decision rule are ignored.
In the present application, when the variable dimension of the full-region decision rule is equal to the variable dimension of the non-full-region decision rule, in the process of generating the full-region decision rule according to the non-full-region decision rule, the full-region decision rule is constructed with parameters corresponding to the variable dimensions of the non-full-region decision rule and the full-region decision rule which are the same.
In the present application, the hash space is a hash table.
For example, in one embodiment, for each aerial task decision unit, the input vector is n-dimensional, and the number of regions divided per dimension is a1, a2, A3.., An, so the input total complexity of the decision unit is a1 × a2 × A3.. An. For example, the dimension of the input vector is { height, velocity, angular dominance }, the number of partitions of the region is { (high, medium, low), (fast, medium, slow), (good, bad) }, and the total complexity is 3 × 2 — 18.
However, when the number of dimensions and the number of regions divided into each dimension are large in the above embodiments, permutation and combination explosion occurs, and manual one-by-one filling cannot be realized.
The present application utilizes the incomplete region partition characteristic of human beings for the fixed rule, for example, the above embodiment has known decision system rules including:
1) if the height is high and the speed is high, the missile is launched,
2) if the height is high, the speed is fast or slow, and the angle advantage is good, the missile is launched,
3) if the height is low and the speed is slow, the missile is not launched,
the three incomplete area decision rules can guide to cover all decision rules, for example, for rule 1, a complete area decision rule can be automatically generated, including that { height is high, speed is high, and angle advantage is good, then a missile is launched; and if the height is high, the speed is high, and the angle advantage is bad, launching the missile }. For rule 2, a full rule may be automatically generated, including launching a missile if the height is high, the speed is fast, and the angle advantage is good; and if the height is high, the speed is slow, and the angle advantage is good, launching the missile }. Thus, a full decision rule can be generated by an incomplete decision rule given by a known decision system.
It can be seen from the above embodiments that the variable dimension of the global decision rule is not less than the variable dimension of the non-global decision rule, and when the variable dimension of the global decision rule is greater than the variable dimension of the non-global decision rule, the parameter corresponding to the variable dimension of the global decision rule greater than the global decision rule is ignored. For example, in rule 1, the angular dominance parameter is ignored. When the variable dimension of the global decision rule is equal to the variable dimension of the non-global decision rule, the global decision rule is constructed by using the parameters corresponding to the variable dimension of the non-global decision rule. E.g., rule 2, then a full-area decision rule is constructed entirely with non-full-area decision rules of height, speed, and angular dominance.
If the rule is not covered with the full decision rule after expansion, a default rule is supplemented, namely the rule which is not covered uniformly fills the default rule.
The variables relevant to air combat are used as inputs to the method in the following descriptions, for example:
the 20-dimensional variables with inputs of x, y, z, velocity, entry angle, …,
the output is { the maneuvering instruction belongs to (flying flat, diving, blocking shooting, breaking S, turning over), the missile launching instruction belongs to (launching, not launching) },
then output permutation combinations of { (fly flat, launch), (dive, no launch), (intercept, launch), (break S, no launch) … } and so on will be generated.
In a decision making system, there are typically multiple decision points, as shown in FIG. 2:
taking an example of an enemy speed situation decision node, the input is { enemy speed e (fast, medium and slow), my party speed e (fast, medium and slow) }, the output is { enemy situation e (poor, medium, good and good) }, and the full permutation and combination is shown in table 1:
TABLE 1
The above is a general inference decision structure based on a hash table, and when the structure is used, the first two columns of key values are encoded, for example: fast & fast, or fast & medium, the reasoning process is: fast & medium- > medium, fast & medium- > good. Since the hash rules are nine pieces in total, the hash rules can be manually filled in.
When the input dimension of a certain decision node is high and each dimension is divided into a plurality of regions, the problem of combination explosion occurs, manual filling cannot be performed, and incomplete rules need to be automatically filled, for example:
and (3) maneuvering selection nodes: the input has three dimensions, each divided into five regions:
{ the friend or foe speed situation is e (poor, medium, good), the friend or foe angle situation is e (poor, medium, good), the friend or foe height situation is e (poor, medium, good);
the output is: { maneuvering instructions belong to (fly flat, dive, intercept, break S, turn to …) }, since the inference structure is based on a hash table, key value encoding needs to be input, according to the permutation and combination principle, the input dimension is 5 × 5 ═ 125 dimensions, the input scale is large, manual filling is difficult, and the following rules are summarized according to cognition, as shown in table 2:
TABLE 2
Friend or foe speed situation | Friend or foe angle situation | Friend or foe height situation | Outputting instructions |
Good or very good | * | Good or very good | Interception jet |
Good or very good or medium | Poor or very poor | * | Is turned into |
Poor or very poor | Poor or very poor | Is very good | Is turned into |
Poor or very poor | Good or very good or medium | * | Crushing of |
Poor or very poor | Good or very good | Poor or very poor | Is rolled out |
Medium or poor or very poor | Good or very good or medium | Good or very good | Dive |
* | * | * | Plane fly |
Table 2 indicates that all regions are included { very bad, medium, good }.
Then rules are automatically generated from the cognitive table, as shown in table 3:
TABLE 3
Friend or foe speed situation | Friend or foe angle situation | Friend or foe height situation | Outputting instructions | Automatic generation of rule numbers |
Good or very good | * | Good or very good | Interception jet | 2*5*2=20 |
Good or very good or medium | Poor or very poor | * | Is turned into | 3*2*5=30 |
Poor or very poor | Poor or very poor | Is very good | Is turned into | 2*2*1=4 |
Poor or very poor | Good or very good or medium | * | Crushing of | 2*3*5=30 |
Poor or very poor | Good or very good | Poor or very poor | Is rolled out | 2*2*2=8 |
Medium or poor or very poor | Good or very good or medium | Good or very good | Dive | 3*3*2=18 |
* | * | * | Plane fly | 5*5*5=125 |
Then, the input is encoded from back to front, the hash table is written according to the encoding and the output, if the encoding is collided, the output is replaced by covering, and all rules can be automatically generated according to cognition.
In addition, as shown in fig. 3, the present application further provides a large-scale air task decision system, wherein the system 10 includes: the knowledge acquisition module 11 is used for acquiring a known air task decision knowledge system; a non-global decision rule generating module 12, configured to construct a non-global decision rule for an air task decision according to a known air task decision knowledge system; a whole region decision rule generating module 13, configured to generate a whole region decision rule according to the non-whole region decision rule and fill the whole region decision rule into a hash space; and the decision module 14 is used for executing the decision of the over-the-air task according to the full-region decision rule in the hash space.
In the present application, the variable dimension of the full-region decision rule is greater than or equal to the variable dimension of the non-full-region decision rule.
In a preferred embodiment of the present application, when the variable dimension of the full-region decision rule is greater than the variable dimension of the non-full-region decision rule, in the process of generating the full-region decision rule according to the non-full-region decision rule, the full-region decision rule is constructed with parameters corresponding to the same variable dimension of the non-full-region decision rule and the full-region decision rule, and the parameters corresponding to the variable dimension of the non-full-region decision rule which is less than the full-region decision rule are ignored.
In a preferred embodiment of the present application, when the variable dimension of the full-region decision rule is equal to the variable dimension of the non-full-region decision rule, in the process of generating the full-region decision rule according to the non-full-region decision rule, the full-region decision rule is constructed with parameters corresponding to the same variable dimension of the non-full-region decision rule and the full-region decision rule.
In a preferred embodiment of the present application, the hash space is a hash table.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A large-scale over-the-air task decision-making method, the method comprising:
acquiring a known air task decision knowledge system;
constructing a non-full-area decision rule of an air task decision according to a known air task decision knowledge system;
generating a full-area decision rule according to the non-full-area decision rule and filling the full-area decision rule into a Hash space;
and executing the decision of the air task according to the whole area decision rule in the hash space.
2. The large-scale over-the-air task decision-making method of claim 1, wherein a variable dimension of the full-area decision rule is greater than or equal to a variable dimension of the non-full-area decision rule.
3. The large-scale air task decision-making method according to claim 2, wherein when the variable dimension of the full-area decision rule is larger than the variable dimension of the non-full-area decision rule, in the generating a full-area decision rule according to the non-full-area decision rule, the full-area decision rule is constructed with parameters corresponding to the same variable dimension of the non-full-area decision rule and the full-area decision rule, and the parameters corresponding to the variable dimension of the non-full-area decision rule smaller than the full-area decision rule are ignored.
4. The large-scale over-the-air task decision-making method according to claim 2, wherein when the variable dimension of the full-area decision rule is equal to the variable dimension of the non-full-area decision rule, in the generating a full-area decision rule according to the non-full-area decision rule, the full-area decision rule is constructed with parameters corresponding to the same variable dimension of the non-full-area decision rule and the full-area decision rule.
5. The large-scale over-the-air task decision-making method of claim 1, wherein the hash space is a hash table.
6. A large-scale over-the-air mission decision system, the system comprising:
the knowledge acquisition module is used for acquiring a known air task decision knowledge system;
the system comprises a non-whole area decision rule generating module, a whole area decision rule generating module and a whole area decision rule generating module, wherein the non-whole area decision rule generating module is used for constructing a non-whole area decision rule of an aerial task decision according to a known aerial task decision knowledge system;
a whole area decision rule generating module, configured to generate a whole area decision rule according to the non-whole area decision rule and fill the whole area decision rule into a hash space;
and the decision module is used for executing the decision of the air task according to the whole area decision rule in the Hash space.
7. The large-scale over-the-air task decision system of claim 6, wherein a variable dimension of the full-area decision rule is greater than or equal to a variable dimension of the non-full-area decision rule.
8. The large-scale air task decision system according to claim 7, wherein when the variable dimension of the full-area decision rule is larger than the variable dimension of the non-full-area decision rule, in the generating a full-area decision rule according to the non-full-area decision rule, the full-area decision rule is constructed with parameters corresponding to the same variable dimension of the non-full-area decision rule and the full-area decision rule, and the parameters corresponding to the variable dimension of the non-full-area decision rule smaller than the full-area decision rule are ignored.
9. The large-scale over-the-air task decision system of claim 7, wherein when a variable dimension of the full-area decision rule is equal to a variable dimension of the non-full-area decision rule, the full-area decision rule is constructed with parameters corresponding to a variable dimension that is the same as the variable dimension of the non-full-area decision rule and the full-area decision rule in the generating the full-area decision rule according to the non-full-area decision rule.
10. The large-scale air task decision system of claim 6, wherein the hash space is a hash table.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911337772.3A CN111125448B (en) | 2019-12-23 | 2019-12-23 | Large-scale aerial task decision method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911337772.3A CN111125448B (en) | 2019-12-23 | 2019-12-23 | Large-scale aerial task decision method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111125448A true CN111125448A (en) | 2020-05-08 |
CN111125448B CN111125448B (en) | 2023-04-07 |
Family
ID=70501258
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911337772.3A Active CN111125448B (en) | 2019-12-23 | 2019-12-23 | Large-scale aerial task decision method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111125448B (en) |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050187847A1 (en) * | 2004-02-20 | 2005-08-25 | Bonissone Piero P. | Systems and methods for multi-objective portfolio analysis and decision-making using visualization techniques |
US20110169835A1 (en) * | 2008-06-20 | 2011-07-14 | Business Inteeligence Solutions Safe B.V. | Dimension reducing visual representation method |
CN105354588A (en) * | 2015-09-28 | 2016-02-24 | 北京邮电大学 | Decision tree construction method |
CN106384282A (en) * | 2016-06-14 | 2017-02-08 | 平安科技(深圳)有限公司 | Method and device for building decision-making model |
CN108985355A (en) * | 2018-06-28 | 2018-12-11 | 中国空间技术研究院 | A kind of data fusion method based on the orthogonal local sensitivity Hash of grouping |
CN109214516A (en) * | 2018-07-20 | 2019-01-15 | 中国航空工业集团公司沈阳飞机设计研究所 | A kind of complete decision making method of fuzzy inference system |
CN109271936A (en) * | 2018-09-18 | 2019-01-25 | 哈尔滨工程大学 | The building of aircraft vehicle vibrations Mishap Database and search method based on perceptual hash algorithm |
CN109991987A (en) * | 2019-04-29 | 2019-07-09 | 北京智行者科技有限公司 | Automatic Pilot decision-making technique and device |
CN110304045A (en) * | 2019-06-25 | 2019-10-08 | 中国科学院自动化研究所 | Intelligent driving transverse direction lane-change decision-making technique, system and device |
-
2019
- 2019-12-23 CN CN201911337772.3A patent/CN111125448B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050187847A1 (en) * | 2004-02-20 | 2005-08-25 | Bonissone Piero P. | Systems and methods for multi-objective portfolio analysis and decision-making using visualization techniques |
US20110169835A1 (en) * | 2008-06-20 | 2011-07-14 | Business Inteeligence Solutions Safe B.V. | Dimension reducing visual representation method |
CN105354588A (en) * | 2015-09-28 | 2016-02-24 | 北京邮电大学 | Decision tree construction method |
CN106384282A (en) * | 2016-06-14 | 2017-02-08 | 平安科技(深圳)有限公司 | Method and device for building decision-making model |
CN108985355A (en) * | 2018-06-28 | 2018-12-11 | 中国空间技术研究院 | A kind of data fusion method based on the orthogonal local sensitivity Hash of grouping |
CN109214516A (en) * | 2018-07-20 | 2019-01-15 | 中国航空工业集团公司沈阳飞机设计研究所 | A kind of complete decision making method of fuzzy inference system |
CN109271936A (en) * | 2018-09-18 | 2019-01-25 | 哈尔滨工程大学 | The building of aircraft vehicle vibrations Mishap Database and search method based on perceptual hash algorithm |
CN109991987A (en) * | 2019-04-29 | 2019-07-09 | 北京智行者科技有限公司 | Automatic Pilot decision-making technique and device |
CN110304045A (en) * | 2019-06-25 | 2019-10-08 | 中国科学院自动化研究所 | Intelligent driving transverse direction lane-change decision-making technique, system and device |
Non-Patent Citations (3)
Title |
---|
QI SHE: ""Distributed High-Dimension Matrix Operation Optimization on Spark"" * |
梁娟: ""基于目标约束优化哈希算法的大规模图像检索研究"" * |
王伟;许云峰;高凯;: "基于哈希表的动态向量降维方法的研究及应用" * |
Also Published As
Publication number | Publication date |
---|---|
CN111125448B (en) | 2023-04-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105867119B (en) | A kind of big envelope curve method for handover control of re-entry space vehicle using protection mapping theory | |
Johansson et al. | Real-time Allocation of Firing Units To Hostile Targets. | |
CN111125448B (en) | Large-scale aerial task decision method and system | |
KR101526893B1 (en) | Simulation method for determining type and requirement quantity of weapons in engagement of air defense and System thereof | |
CN109214043B (en) | Artificial intelligence writing method for digital aircraft dynamics environment information transmission source code | |
CN109189376B (en) | Artificial intelligence writing method for digital aircraft cluster source code | |
Tabassum | Integrated approach of set theory and pattern search methods for solving aircraft attacking problem | |
Daughtery et al. | Optimal design of cooperative guidance law for simultaneous strike | |
US9830408B1 (en) | System and method for evaluating the performance of a weapon system | |
CN114528750A (en) | Intelligent air combat imitation learning training sample generation method based on self-game model | |
CN115222023A (en) | Neural network-based non-cooperative target control strategy identification method and system | |
Kim et al. | New weapon target assignment algorithms for multiple targets using a rotational strategy and clustering approach | |
CN109214516B (en) | Complete decision generation method of fuzzy inference system | |
Hu et al. | Research on evolution method of aircraft swarms space configuration based on event-driven | |
Polak et al. | Method of outer approximations and adaptive approximations for a class of matrix games | |
CN113705102B (en) | Deduction simulation system, deduction simulation method, deduction simulation equipment and deduction simulation storage medium for sea-air cluster countermeasure | |
Zhao et al. | Mission-oriented scheme generation method for weapon system of systems | |
Djiknavorian et al. | Approximation in DSm theory for fusing ESM reports | |
McCormick et al. | Artificial intelligence to assist weapon package selection for airborne weapon delivery | |
CN111817770B (en) | Relay unmanned aerial vehicle deployment method and terminal equipment | |
CN108199868A (en) | A kind of group system distributed control method based on tactics cloud | |
Preston | Interdiction of a transportation network | |
Yu et al. | Application of Bayesian for the Situation Assessment of Sea-battlefield | |
CN116466736A (en) | Air combat dog bucket rolling maneuver auxiliary decision making method based on decision tree | |
Lei et al. | Air Combat Intelligent Assistant Decision-Making Sample Data Cultivation Method |
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 |