Aircraft angle-of-attack virtual sensor design via a functional pooling narx methodology

The design of an Angle-of-Attack aircraft Virtual Sensor is pursued via a novel Functional Pooling Nonlinear AutoRegressive with eXogenous excitation (FP-NARX) methodology. This methodology essentially is an identification approach capable of establishing a nonlinear dynamical system model from data obtained from many different flights, each one corresponding to different aircraft and environmental conditions. The FP-NARX model structure is shown to be suitable for Angle-of-Attack Virtual Sensor development within the landing, takeoff, and clean flight regimes of the aircraft. Its performance is examined via validation flights, which include substantial aircraft maneuvering, and is shown to be very good, with peak errors not exceeding 1.1 degrees.