Distributed estimation for time-varying target in noisy environment

This paper studies the continuous-time distributed estimation problem for time-varying target under switching topologies and stochastic noises. There are three main features in this problem: only a portion of sensors have a access to the target; three kinds of stochastic noises arising in dynamic process, measurement and communication are considered; and the topological structure between sensors and target is switching. For this problem we propose a continuous-time distributed estimation algorithm. Under observability and connectivity, one upper and lower bound for the total mean square estimation error is established by using common Lyapunov method and Kalman-Bucy filtering theory, respectively. The numerical simulation also verifies the effictiveness of the proposed algorithm.