Agent-based centralized fuzzy Kalman filtering for uncertain stochastic estimation

In this paper, we investigate the problem of agent-based centralized fuzzy Kalman filters observing an uncertain physical process with parametric uncertainties. An agent-based sensor network is a distributed system which consists of sensors with limited computational capabilities. In our agent-based sensor network we consider sensor agents and a moderator agent. Any of these sensor agents have limited computational capabilities and also may be affected by different noises. Agents derive the information in the form of fuzzy states from their fuzzy Kalman filters, the estimated fuzzy states would be transmitted to the moderator agent for aggregation and result sharing by any sensor agent. The moderator agent fuses the fuzzy estimations to generate the global state estimations which is highly reliable.