Unscented Kalman Filtering for Reentry Vehicle Identification in the Transonic Regime

Abstract

Parameter identification methods for processing flight data are frequently used to validate and improve a preflight aerodynamic database and, specifically, to reduce the associated uncertainties. In this framework, the paper describes an identification methodology developed for the first flying test bed of the Italian Aerospace Research Center, a demonstrator of technologies relevant to future reusable launch vehicles. The analysis is focused on aerodynamic modeling of the reentry vehicle configuration in the transonic flow regime, in which flight control system performance is affected by a significant level of parameter uncertainty. The parameter estimation is formulated as a nonlinear filtering problem and solved through a multistep approach, in which the aerodynamic coefficients are identified first and, in a following phase, a set of model parameters is updated. In each step, an unscented Kalman filter is used as a recursive estimation algorithm. The methodology is applied to the flight data of the Dropped Transonic Flight Test mission of the vehicle, carried out during the winter of 2007. The reported results demonstrate the good characteristics of the technique in terms of convergence, reduction of uncertainty of the a priori aerodynamic model, and capability of extracting the information content from a rather limited set of flight data on vehicle response.