Distributed Filtering and Estimation for Industrial Systems

Abstract

Distributed Filtering and estimation methods for industrial systems are studied. Such methods are particularly useful in case that measurements about the industrial system are collected and processed by n different monitoring stations. The overall concept is that at each monitoring station a filter is used to track the state of the system by fusing measurements which are provided by various sensors, while by fusing the state estimates from the distributed local filters an aggregate state estimate for the industrial system is obtained. In particular, the chapter proposes first the Extended Information Filter (EIF) and the Unscented Information Filter (UIF) as possible approaches for fusing the state estimates provided by the local monitoring stations, under the assumption of Gaussian noises. The EIF and UIF estimated state vector can, in turn, be used by nonlinear controllers which can make the system’s state track desirable setpoints. Moreover, the Distributed Particle Filter (DPF) is proposed for fusing the state estimates provided by the local monitoring stations (local filters). The motivation for using DPF is that it is well-suited to accommodate non-Gaussian measurements. The DPF estimated state vector can again be used by a nonlinear controller to make system converge to desirable setpoints. The performance of the Extended Information Filter, of the Unscented Information Filter and of the Distributed Particle Filter is evaluated through simulation experiments in the case of a 2-UAV (unmanned aerial vehicle) model monitored and remotely navigated by two local stations.