Multiagent approach for aerospace attack scenarios description

Аuthors

Abrosimov V. K.^1*, Goncharenko V. I.^2**

1. "Step Logic" Company, 3, Polkovaya str., building 3, Moscow, 127018, Russia
2. Moscow Aviation Institute (National Research University), 4, Volokolamskoe shosse, Moscow, А-80, GSP-3, 125993, Russia

*e-mail: avk787@hotmail.com
**e-mail: fvo@mai.ru, vladimirgonch@mail.ru

Abstract

The article is dedicated to potential enemy possibilities evaluation, when one develops aerospace defense system.
It is considered the new hypothesis, which supposes an opportunity of data exchange between attacking flying vehicles. As the result the main core of controlled aerospace attack is distributed intellectual control system.
It is suggested to use so called «multi agent approach» for described above event simulation. The set of agent states at each simulation level is represented as finite automaton, which has corresponding transfer functions. So, one receives the new opportunities to manage attacking vehicles maneuvers in order to create an additional complexity to aerospace defense system operation.
Attacking flying vehicles are represented as interacting intellectual agents with Ai state vector, where
Ai =<Bi, Gi, Si, Ni, UiEij, > ,i=1, n,
here are: Bi is knowledge base of i-agent;
Gi is set of i-agent targets;
Si is set of i-agent strategies;
Ni is set of i-agent intentions;
Ui is set of i-agent obligations;
E is multidimantional matrix, which defines «i» and «j» agents mutual interconnections, considering agents state, targets, intentions etc.
The set of i-agent targets is defined by given agent set of tasks. The set of i-agent strategies defines the set of given agent program trajectories. The set of i-agent intentions is defined by set of targets, sel ected fr om the point of view given agent efficiency. The set of i-agent obligations is generated as the result of data exchange between different attacking flying vehicles. In particular, these obligations have to include the readiness to attack target, which can not be attacked by others agents.
So, the main innovative performance of such aerospace attack is collective intellectual behavior of attacking flying vehicles.
The basic concept of aerospace attack scenarios description is developed. The simulation of aerospace attack scenarios is provided by neural network utilization, because such approach gives an opportunity to implement simultaneous computation of various attacking vehicles trajectories.

Keywords:

anti-aircraft defence, anti-ballictic missile defense, aerospace attack scenario, agent-oriented approach, intellectual agent, strategy, modeling

References

1. Molitvin A. Zarubezhnoe voennoe obozrenie, 2008, no. 1, pp. 23-27.
2. Sergeev I.D., Yakovlev V.N., Solovtsov N.E. Voennyi entsiklopedicheskii slovar Raketnykh voisk strategicheskogo naznacheniya (Strategic Missile Forces), Мoscow, Bolshaya Rossiiskaya entsiklopediya, 1998, 632 p.
3. Fateev V.F., Sukhanov S.A., Omelchuk V.P. Vozdushno-kosmicheskaya oborona, 2006, no. 4(29), pp. 2-7.
4. Tarasov V.B. Ot mnogoagentnykh sistem k intellektualnym organizatsiyam (From multi-agent systems to intellectual commuhities), Мoscow, Editorial URSS, 2002, 352 p.
5. Pospelov D.A. Materialy mezhdunarodnogo seminara «Raspredelennyi iskusstvennyi intellekt i multiagentnye sistemy», DIAMAS - 97, St.-Petersburg, 1997, pp. 319-325.
6. Zhukov S.Yu. Navigatsiya intellektualnykh agentov v slozhnykh sinteticheskikh prostranstvakh (Navigation of intellectual agents in complex synthetic spaces), PhD thesis, St.-Petersburg, 2000, 135 p.
7. Shvetsov A.N. Agentno-orientirovannye sistemy: ot formalnykh modelei k promyshlennym prilozheniyam, 2008, available at: http://www.ict.edu.ru/lib/index. php?id_res=5656
8. Bolshaya tekhnicheskaya entsiklopediya (Big technical encyclopedia), Мoscow, Nauka, 1994, vol. 6, 873 p.
9. Ivashkin Yu.A. Materialy mezhdunarodnoi konferentsii «Parallelnye vychisleniya i zadachi upravleniya», Мoscow, IPU RAN, 2001, pp. 132-137.
10. Averkin A.N., Blishun A.F., Batyrshin I.Z., Silov V.B., Tarasov V.B. Nechetkie mnozhestva v modelyakh upravleniya i iskusstvennogo intellekta (Fuzzy sets technigue upplying to both control and artificial intellect models), Мoscow, Nauka, 1986, 312 p.
11. Programmnoe obespechenie dlya modelirovaniya, analiza i vizualizatsii kosmicheskikh, voennykh i razvedyvatelnykh sistem http://www.agi.com/default.aspx
12. Karpov Yu. Imitatsionnoe modelirovanie system (Systems modeling simulation), St.-Petersburg, BVKh-Peterburg, 2005, 400 p.