Estimation and Control of Large-Scale Networked Systems

Chapter 6

Distributed State Estimation in an LSS

Abstract

In this chapter, we discuss design of a distributed observer which has the same structure as that of the plant. Under the requirement of unbiasedness, we derive an explicit form for the optimal gain that minimizes the covariance matrix of estimation errors. Asymptotic properties of this observer are also investigated. It has been made clear that in order to have the same estimation accuracy as that of the lumped Kalman filter in the steady state, the update gain of the Kalman filter must converge to a block diagonal matrix.

Keywords

covariance minimization; distributed estimation; recursive estimation; unbiased estimate

6.1 Introduction

In various engineering applications, ...