Unscented and Extended Kalman Estimators for non Linear Indoor Tracking Using Distance Measurements

Abstract

Industrial and logistic indoor tracking with accuracy in centimetre range is still a challenging issue. Many applications in mining, logistic, and navigation depend mainly on a precise determination of a mobile terminal. This work presents an indoor positioning approach using two recursive tracking algorithms for precisely localizing a mobile vehicle in a noisy environment. An extended Kalman filter (EKF) and unscented Kalman filter (UKF), the corresponding algorithms and mathematical models are presented and analysed. Experimental range measurements obtained from local positioning radar system are used to feed the filters. True and estimated trajectories of the mobile vehicle with associated means and error covariances are presented in more details. Results obtained shows that UKF is slightly more accurate and reliable. Whereas, EKF is easier to implement, converges faster when fed with a good initial estimate and more optimized for semi-linear tracking models.