Article

Article title NORMATIVE MULTI-AGENT SYSTEMS
Authors A.I. Mikov
Section SECTION III. SOFTWARE ENGINEERING
Month, Year 06, 2014 @en
Index UDC 004.436.4; 510.643
DOI
Abstract The concept of normative agent"s goals and action constraints are formulated in the language of deontic logic. Agent is guided by external norms (regulations), describing the environ- ment (subject area) and opportunities and prohibitions in conjunction with the agent environment. Normative multi-agent systems represent a set of interacting regulatory agents that solve a common problem in a distributed system. Norms may form a hierarchical regulatory system. Somekinds of norms are distinguished such as regulatory, declarative and procedural. Relationships between agents in multi-agent system are also presented in the form of regulations. The role of the “count as” relation in normative statements is discussed. This paper describes some different types of deontic operators for normative agents, variants of agent actions, statements of action status, the set which are deontic closed by actions. Also the paper discusses cooperation of agents, coalitions, the problem of formalizing the rules specified in natural language, restrictions on the behavior of agents in time using the model of branching time. The examples of normative texts transformation into formulas of applied modal logics are given. The set-theoretic definitions of normative systems and their description by formal grammars are compared.

Download PDF

Keywords Deontic logic; norm; program agent; descriptive language.
References 1. Milner R. Elements of interaction. Turing Award lecture // Communication of the ACM. – 1993. – Vol. 36, № 1. – P. 78-89.
2. Milner R., Parrow J., Walker D. Modal logics for mobile processes // Proceedings of 2nd International Conference on Concurrency Theory (Amsterdam, The Netherlands, August 26–29, 1991 (CONCUR '91). – Lecture Notes in Computer Science. – 1991. – Vol. 527. – P. 45-60.
3. Миков А.И. Информационные процессы и нормативные системы в IT: Математические модели. Проблемы проектирования. Новые подходы. – М.: Книжный дом «ЛИБРОКОМ», 2013. – 256 с.
4. Jones A., Carmo J. Deontic logic and contrary-to duties // Handbook of Philosophical Logic (D. Gabbay, ed.). – Kluwer, 2001. – P. 203-279.
5. Миков А.И., Воробьев В.В. Моделирование агентов с деонтической логикой, функционирующих в распределенных системах // Информатизация и связь. – 2012. – № 5. – С. 82-85.
6. Миков А.И. Модели сложности разделения и интеграции информационных систем // Информатизация и связь. – 2013. – № 2. – С. 92–96.
7. McNamara P. Agential obligation as non-agential personal obligation plus agency // Journal of Applied Logic. – 2004. – Vol. 2. – P. 117-152.
8. Eiter T., Subrahmanian V.S., Rogers T.J. Heterogeneous active agents, III: Polynomially implementable agents // Artificial Intelligence. – 2000. Vol. 117. – P. 107-167.
9. Миков А.И. Аксиоматика отношений в предметных областях и реляционные отношения // Информатизация и связь. – 2013. – № 5. – С. 70-74.
10. Wooldridge M., van der Hoek W. On obligations and normative ability: Towards a logical analysis of the social contract // Journal of Applied Logic. – 2005. – Vol. 3. – P. 396-420.
11. Horty J.F. Agency and Deontic logic. – Oxford: Oxford University Press, 2001.
12. Boella G., van der Torre L. Substantive and Procedural Norms in Normative Multiagent Systems // Journal of Applied Logic. – 2008. Vol. 6. – P. 152-171.
13. Castelfranchi C. Modeling social action for AI agents // Artificial Intelligence. – 1998. – Vol. 103, № 1-2. – P. 157-182.
14. Felicнssimo C., de Lucena C.J. P., Briot J.-P. A Norm-Based Approach for the Modeling of Open Multiagent Systems // Int. Conf. on Agents and Artificial Intelligence (Porto, Portugal, ICAART‟09). – P. 540-546.
15. Garcia-Camino A., Noriega P., Rodriguez-Aguilar J.A. Implementing norms in electronic institutions // Proceedings of the 4th international joint conference on Autonomous agents and multiagent systems (The Netherlands). – 2005. – P. 667-673.

Comments are closed.