Improvement of bearings only target tracking using smoothing

Abstract

Bearings only target tracking is often addressed using linearized estimators such as the extended Kalman filter (EKF). Due to the erratic performance of the EKF algorithm in passive localization, a new filtering method, referred to as the smoothing modified gain EKF (sMGEKF), is proposed based on the modified gain EKF (MGEKF) and Rauch-Tung-Striebel (RTS) smoothing. Compared to the traditional approaches (e.g., the EKF and the iterated EKF) used in passive localization, the proposed method has potential advantages in tracking accuracy. A simulation study demonstrates the advantages of this approach.