Online Estimation of Covariance Parameters using Extended Kalman Filtering and Application to Robot Localization

Abstract

Abstract This paper presents a novel method for the online estimation of variance parameters regulating the dynamics of a nonlinear dynamic system. The approach exploits and extends classical iterated Kalman filtering equations by propagating an approximation of the marginal posterior of the unknown variances over time. In addition to the theoretical foundations, this manuscript offers also a variety of numerical results. In particular, experiments with data collected both in simulation and with a real robot platform show how the proposed approach efficiently solves a robot localization problem.