Stochastic stability of decentralized Kalman filter for nonlinear systems

Abstract

In this paper, stochastic stability of a decentralized Kalman filter (DKF) for nonlinear systems and a bound on the sum of square of predicted estimation error are examined of a single node. A Lyapunov analysis approach is used to show that the predicted estimation error of the nonlinear DKF is stochastically stable and exponentially bounded in mean square, if the system is observable, controllable and the initial estimation error is bounded. Moreover, it is validated by the numerical simulations. The numerical results show that the sum of square of the predicted estimation error is bounded above.

Hierarchical recursive least squares parameter estimation methods for multiple‐input multiple‐output systems by using the auxiliary models

“…Therefore, the key of identifying MIMO systems is to improve the calculation efficiency. The hierarchical identification principle is an effective tool [8][9][10] to solve the parameter identification issue of large-scale systems with the heavy computational burden and can be applied to multivariable systems, 11,12 nonlinear systems, [13][14][15] and bilinear systems. 16 The hierarchical identification principle is divided into three steps: the identification model decomposition, the submodel identification, and the coordination of correlation terms among sub-algorithms.…”

Finite‐time composite control for fuzzy delayed semi‐Markovian jump systems with multiple disturbances and uncertainty

“…Authors of Reference 11 designs an estimator to estimate the state of the nonlinear systems to ensure the system performance. Authors of Reference 12 discussed the stability problem of nonlinear systems with Kalman filter. Output feedback control and application for nonlinear system has been discussed in Reference 13.…”