Dynamics of approximate information fusion
Linköping University, Department of Computer and Information Science, KPLAB - Knowledge Processing Lab
Linköping University, The Institute of Technology
Conference paper (Refereed)
Proceedings of the International Conference on Rough Sets and Emerging Intelligent Systems Paradigms (RSEISP)
Place of publ.:
Lecture Notes in Artificial Intelligence, ISSN 0302-9743; 4585
The multi-agent system paradigm has proven to be a useful means of abstraction when considering distributed systems with interacting components. It is often the case that each component may be viewed as an intelligent agent with specific and often limited perceptual capabilities. It is also the case that these agent components may be used as information sources and such sources may be aggregated to provide global information about particular states, situations or activities in the embedding environment. This paper investigates a framework for information fusion based on the use of generalizations of rough set theory and the use of dynamic logic as a basis for aggregating similarity relations among objects where the similarity relations represent individual agents perceptual capabilities or limitations. As an added benefit, it is shown how this idea may also be integrated into description logics.