Performance Comparison of EKF and UKF for Estimation of COE using Kozai Mechanism

Abstract

The paper presents performance comparison of Extended Kalman Filter and Unscented Kalman Filter for estimation of Classical Orbital Elements (COE) using RADAR measurements. SGP4 algorithm has been used for relating the Two Line Elements (TLE) information with orbital parameters and subsequently is used for generating the RADAR measurements of position and velocity in the orbit. Zero mean, white Gaussian additive noise has been considered for RADAR measurements. The nonlinearity in the state and measurement models have been dealt with EKF and UKF for estimation of TLE information at every epoch and the results of both the estimators have been compared.