Store-induced limit cycle oscillations (LCOs) are a common issue on current fighter aircraft and are expected to be present on next generation fighter aircraft. Current efforts in control systems designed to suppress LCO behavior have been either linear and restricted to specific flight regimes or nonlinear but require uncertainties in the system dynamics to be linear-in-the-parameters and only present in the torsional stiffness. Furthermore, the aerodynamic model used in prior research efforts neglects any nonlinear effects. This paper presents the development of a controller consisting of a continuous RISE feedback term with a neural network feedforward term to achieve semiglobal asymptotic tracking of the wing angle of attack in the presence of structural and aerodynamic uncertainties that do not satisfy the linear-in-the-parameter assumption.

This content is only available via PDF.
You do not currently have access to this content.