A Comprehensive Flight Data Based Model Prediction : Perspective Analysis and Comparison

Abstract

Unmanned High Speed Aerial Vehicle (UHSAV) are an important tool for various applications applications. To ensure mission success, robust control system needs to be developed for these UHSAVs, which require well characterized dynamic system model. This paper aims on model estimation of an experimental UHSAV utilizing actual flight data. An elaborate estimation mechanism is proposed utilizing various model structures techniques such as Autoregressive Exogenous (ARX), Autoregressive Moving Average exogenous (ARMAX), Box Jenkin’s (BJ), Output Error (OE), and state space and non-linear Autoregressive Exogenous. A perspective analysis and comparison is made for identifying the salient aspects of each individual model structure. Model configuration with best characteristics is then identified based upon model quality parameters such as residual analysis, final prediction error and fit percentages. Extensive validation to evaluate the performance of developed model is then performed utilizing the actual flight dynamics data collected from different flights of the same UHSAV. Results indicate the viability of the model as the model can accurately predict the system performance at a wide range of operating conditions. Through this, to the very of our knowledge we present a comprehensive model prediction framework, which utilizes actual flight data instead of simulation work