Parameters estimation methodology for the nonlinear rolling motion of finned cylindrical body

Abstract

Identification of nonlinear roll dynamics of finned cylindrical bodies is a critical step when assessing free motion stability and trajectories of aerially dispensed munitions or decoys. In this paper the authors present a parameter estimation process that focuses on identifying nonlinear aerodynamic models that characterize the roll dynamics of a cylindrical body with wrap around fins using data from a series of dynamic wind tunnel tests. This is a three step approach that combines ordinary least squares, stepwise regression and the augmented output-error method, and it is initially tested using simulation data corrupted by white Gaussian noise and then applied to the wind tunnel data. Roll and roll rate dynamics were captured through a series of high angle of attack free-to-roll tests carried out at an airspeed of 35m/s corresponding to a Reynolds number of 800,000. The results and discussion in this paper demonstrate how simulation can be used to develop and mature a system identification routine followed by its assessment through wind tunnel test data. It is shown that high order nonlinear models with up to 14 terms can be parameterized to provide high levels of agreement with roll and roll rate dynamics observed in the dynamic wind tunnel tests.