Aerodynamic parameter estimation of a supersonic missile with rapid speed variation by using kalman filtering

---

Tezin Türü: Yüksek Lisans

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mühendislik Fakültesi, Havacılık ve Uzay Mühendisliği Bölümü, Türkiye

Tezin Onay Tarihi: 2016

Öğrenci: TUĞBA BAYOĞLU

Danışman: ALİ TÜRKER KUTAY

Özet:

This study aims to develop an approach for the aerodynamic parameter estimation problem for supersonic air to air missile with rapid speed variation, as well as large variation of aerodynamic parameters with respect to Mach number. In literature, most of the estimation techniques require that estimator is time independent. Therefore, most of the aerodynamic parameter estimation methods are applied for air vehicles which have control over their speed or which can operate at relatively constant Mach number so that the variation of aerodynamic coefficients with respect to Mach number is eliminated. For aircrafts, it is possible to identify aerodynamic parameters during trimmed flight condition at an approximately constant Mach number. Moreover, since a fixed wing aircraft has control over its thrust and speed, many flight tests can be conducted to identify aerodynamic derivatives at discrete Mach numbers. On the contrary, most agile missile configurations do not have control over their speed, number of flight test trials is very limited and flight durations are much less than that of fixed wing aircraft. Therefore, the missile aerodynamic parameter identification problem requires a different approach and this study serves as a guide to these considerations. In this study, aerodynamic parameter identification problem of a supersonic missile, which has large and rapid Mach variations throughout its limited flight duration, is solved by using the Kalman filter algorithm together with measurements and flight path reconstruction outputs modified by the Wiener filter and a complementary filter. First, the mathematical structure of the n onlinear aerodynamic model in the identification region is obtained and the aerodynamic identification problem is simplified to a classical, time-invariant aerodynamic derivative identification problem. Then, Mach derivatives of aerodynamic terms are added into the simplified aerodynamic model to take into account the dependency of the aerodynamic parameters on Mach number. The resulting parameter identification problem is solved by using an adaptive Kalman filter algorithm. Finally, the suggested method is tested against a set of uncertainties by using simulated data generated from a high fidelity six degree of freedom missile flight simulation model. Since estimation method is vulnerable to drift in angle of attack, a first order complementary filter which is designed using a predefined but uncertain aerodynamic model is utilized to remedy angle of attack reconstruction problems due to instrumentation errors.