Agent Teaming and Learning in Artificial Intelligent Environment
Intelligent Agents (IA) are described in the literature as autonomous computational entities that can respond to the changes in their environment and plan a series of tasks accordingly to achieve their goals. Multi-Agent System (MAS) consist of IA that can interact, communicate, and cooperate with each other. This research aims to develop an interoperable MAS and agent teaming architecture that enables their entities to automatically communicate and adapt their functionality at runtime based on message flows [1][2]. Rigid design-time constraints can be replaced by a flexible plug-and-play componentized capability. Intelligent Agents must possess interoperability and capability to share knowledge and context in order to achieve their goals. A concept demonstrator is being developed, using a number of dynamic distributed environments to show how interoperable Multi-Agent Systems can improve data flow in a distributed environment. The agents in the MAS are equipped with a number of sensors that provide data from the environment which are fused to produce knowledge. The fused information is fed into an inference engine which contains the Subject Mater Expert equipped with knowledge required to make decision and/or change some course of action.
References
[1] Khazab, M., Tweedale, J. and Jain, L.C., Interoperable Intelligent Agents in a Dynamic Environment, New Advances in Decision Technologies, Proceedings of the First International Symposium on Intelligent Decision Technologies, Himeji Japan, April 2009, Springer-Verlag, 2009, pp. 183-191.
[2] Khazab, M., Tweedale, J. and Jain, L.C., Dynamic Applications using Multi-agent Systems, in Intelligent Systems and Technologies, Chapter 4, Springer-Verlag, 2009, pp. 67-79.