Bearing only tracking using Gauss-Hermite filter

Abstract

In this paper, performance of Gauss-Hermite filter (GHF) in bearing only tracking problem has been compared with that of extended Kalman filter (EKF) and unscented Kalman filter (UKF) in terms of estimation accuracy, probability of track-loss and computational efficiency. The performance improvement of the GHF with increase in quadrature points and enhanced robustness compared to EKF and UKF with respect to large initial uncertainty has been reported. It has been concluded that without introducing substantial computational burden, GHF with three or more quadrature points exhibits better performance compared to UKF and EKF.