Networked strong tracking unscented Kalman filter with multiple packet dropouts and parameter perturbations

Abstract

In this paper, a recursive networked strong tracking unscented Kalman filter (NSTUF) for a class of nonlinear networked control systems (NCSs) with parameter perturbations, unknown inputs and network-induced packet dropouts is proposed. Due to the limited network bandwidth, the data transmitted via a shared communication channel from different sensors distributed in a large geographical area may suffer from independent packet dropouts. Then, a series of Bernoulli sequences are employed to describe the multiple packet dropout rates based on which a networked unscented Kalman filter with multiple packet dropouts is first derived. Next, a defined suboptimal fading factor is proposed and added to the covariance of prediction to deal with parameter perturbations and unknown inputs. Therefore, a networked strong tracking unscented Kalman filter is proposed. Finally, a simulation example of the three-tank system is given to show the effectiveness of the proposed method.