Distributed State Estimation for Stochastic Linear Hybrid Systems

Abstract

This paper studies the problem of distributed state estimation for stochastic linear hybrid systems. Building on the centralized interacting multiple model algorithm, a novel distributed state estimation technique is proposed. In our distributed setting, a network of sensors is employed and each sensor measures only a portion of the system outputs. It should be noted that the system might not be observable for each individual sensor. In this paper, we aim to develop an effective scheme that enables the sensor network to collectively estimate the hybrid system states (both the continuous states and discrete modes), by leveraging the cooperation among multiple sensors. Consequently, each sensor only needs to process a relatively small set of data and will be able to locally and identically observe the states of stochastic linear hybrid systems. The validation and performance of the proposed scheme are demonstrated by numerical simulations on an aircraft tracking problem finally.