Unscented Kalman Filter with Application to Bearings-only Passive Target Tracking

Abstract

Background/Objectives: Target tracking is an age old problem which demands robust statistical estimators which can effectively track the target within the acceptable limits of errors in target motion parameters. The objective of this paper is to develop a novel estimation algorithm based target tracking simulator for underwater target tracking applications. Methods/Statistical Analysis: The unscented transformation developed by Julier, et al., is applied to the body of Kalman filter to synthesize Unscented Kalman filter. The modeling of target state and measurement vectors is carried out. Unscented Kalman filter is integrated into the model to result in evolution of simulator. Extensive performance evaluation of UKF with respect to bearings-only target tracking problem in Monte-Carlo simulation is carried out and the results are presented. Findings: UKF algorithms effectively track the target with encouraging convergence time which is proved from the results obtained in single run and Monte-Carlo simulation. It is observed that UKF is suitable algorithm for bearings-only target tracking problem. Application/Improvements: The results obtained are satisfactory and UKF can be used in futuristic submarines in Indian Navy owing to its advantages as envisaged in this paper.