Accurate distributed estimation of the position of network nodes is essential for many applications, including localization, geographic routing, and vehicular networks. When nodes are mobile and their mobility pattern is unknown, there are not yet adequate techniques to achieve high accuracy and low estimation errors. In this paper, a new distributed estimator of the position of mobile nodes is proposed. No model of the mobility is assumed. The estimator combines heterogeneous information coming from pre-existing ranging, speed, and angular measurements, which is jointly fused by an optimization problem where the squared mean and variance of the localization error is minimized. Challenges of this optimization are the characterization of the moments of the noises that affect the measurements. The estimator is distributed in that it requires only local processing and communication among the nodes of the network. Numerical results show that the proposed estimator outperforms traditional approaches based on the extended Kalman filter. © 2011 IFAC.
A sensor fusion algorithm for mobile node localization
DE ANGELIS, ALESSIO;
2011
Abstract
Accurate distributed estimation of the position of network nodes is essential for many applications, including localization, geographic routing, and vehicular networks. When nodes are mobile and their mobility pattern is unknown, there are not yet adequate techniques to achieve high accuracy and low estimation errors. In this paper, a new distributed estimator of the position of mobile nodes is proposed. No model of the mobility is assumed. The estimator combines heterogeneous information coming from pre-existing ranging, speed, and angular measurements, which is jointly fused by an optimization problem where the squared mean and variance of the localization error is minimized. Challenges of this optimization are the characterization of the moments of the noises that affect the measurements. The estimator is distributed in that it requires only local processing and communication among the nodes of the network. Numerical results show that the proposed estimator outperforms traditional approaches based on the extended Kalman filter. © 2011 IFAC.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
