Distributed nonlinear estimation for diverse sensor devices

Distributed linear estimation theory has received increased attention in recent years due to several promising, mainly industrial applications. Distributed nonlinear estimation, however, is still a relatively unexplored field despite the need for such a theory in numerous practical problems with inherent nonlinearities. This work presents a unified way of describing distributed implementations of three commonly used nonlinear estimators: the extended Kalman filter (EKF), the unscented Kalman filter (UKF) and the particle filter. Leveraging on the presented framework, we propose new distributed versions of these methods, in which the nonlinearities are locally managed by the various sensors, whereas the different estimates are merged based on a weighted average consensus process.We show how the merging mechanism can handle sensors running different filters, which is especially useful when they are endowed with diverse local computational capabilities. Numerical simulations of the proposed algorithms are shown to outperform the few published ones in a localization problem via range-only measurements. Quality and effectiveness are investigated in a heterogeneous filtering scenario as well. As a special case, we also present a way to manage the computational load of distributed particle filters using graphical processing unit (GPU) architectures.