Estimation of Ballistic Coefficient of Reentry Vehicle with Divided Difference Filtering using Noisy RF Seeker Data

Abstract

Estimation of relative kinematic states of the re-entry vehicle from noisy seeker measurements is a nonlinear filtering problem. In case of nonlinear filtering problem, choice of any estimator is scenario dependent. Recently few nonlinear estimation techniques such as Unscented Kalman filter (UKF), Divided Difference Filter (DDF) and other techniques promise to be performing some what better than Extended Kalman Filters (EKF), although the claim depends on particular non-linear problem. In this paper, application of Divided Difference Filter is examined in estimating relative kinematic parameters of reentry vehicle. The application of the proposed estimator on noisy measurement data available from seeker is demonstrated and comparison results are shown along with EKF and UKF. These estimators becomes more accurate than estimators based on Taylor approximation like EKF. Basic DDF implementation takes very high computational time because of many state propagation equation are solved online. For present problem, special numerical solution is presented, which makes DDF computation almost as fast as that of EKF for a chosen number of states. This is a new solution to present problem and can be utilized in new genre of filtering with present problem.