System identification of UAV under an autopilot trajectory using ARX and Hammerstein-Wiener methods

The need of an efficient mathematical model that characterizes the Aircraft system dynamics is essential for implementing an autopilot system, control algorithms, and other applications. The challenge of system identification is to predict the parameters of the system and construct a model that can estimate the system behavior under both normal inputs and disturbance. One of the main problems in system identifications is defining the input signal for the system. For aircrafts system identification was done using a test signal that take into consideration the allowed range of the inputs (usually Euler angles and their rates). The main contribution in this paper is in giving more realistic inputs using the adaptive autopilot that control the aircraft trajectory In this paper, system identification of lateral and longitudinal dynamics will be presented for an ARF-60 UAV using both linear and nonlinear estimation methods. The estimation methods are Autoregressive with exogenous model (ARX) method for linear estimation, and Hammerstein-Wiener method for nonlinear estimation. Moreover, a comparison will be made between the model and original system. The input trajectory of the system defines how strong the estimation will be; for example, a straight line trajectory will not give enough data to construct the estimated model. The orders of each estimation is given and its coefficients.