Decentralized navigational state estimation for networked navigation systems with finite channel capacity and randomly switching topologies

Abstract

This article deals with the decentralized navigational state estimation problem for a class of networked navigation systems, in order that the average navigation performance of networked navigation system carriers is guaranteed over a given finite-horizon. The observations of the novel system, comprising the cooperative information of the neighbors and the locally relative measures, are transmitted through a randomly switching topology network with finite channel capacity, and thus are exposed to the impacts of quantization effect, channel saturation, and colored noise. In this study, the above issues are tackled via a decentralized receding horizon fusion scheme, which comprises two stages. At the first stage, every networked navigation system collects the sink knowledge from its neighbors to generate a local estimate, and then the remote-access cooperative estimates, with non-equal horizon length, are further fused at the second stage to form a global estimate. Local optimal estimators are designed based on receding horizon Finite Impulse Response (FIR) estimation principle; while in global estimator, a fusion criterion for error cross-covariance is derived between local and cooperative estimates with different horizon lengths. A numerical simulation example is provided to demonstrate the effectiveness and applicability of the developed estimation method.