The reference ontology of collective behavior of autonomous agents and its extensions
V. I. Gorodetsky, Vladimir Samoylov, D. V. Trotskii · Journal of Computer and Systems Sciences International · 2015
A behavioral paradigm of artificial intelligence (AI) systems is considered. In this paradigm, it is assumed that the system’s “intelligence” emerges as a result of the individual behaviors and interaction of a set of distributed entities (robots, software agents, and the like) between themselves and with the external environment. An outline of the state of art in the field of behavioral models of artificial intelligence systems is given and, for such models, a unified semantically interpreted behavioral metamodel in the form of a domain-independent reference ontology and its extensions for two particular practically important classes of applications are proposed. The first class of applications deals with the team work of underwater robots autonomously inspecting underwater space in order to ensure its security. The second one corresponds to self-organizing systems composed of a large number of small satellites that autonomously communicate, observe, and inspect outer space. Directions of future research in the field of behavioral models of the distributed entities that cooperatively accomplish an autonomous mission are outlined.