Aircraft System Identification Using Artificial Neural Networks
Conference Paper
Overview
Identity
Other
View All
Overview
abstract
This paper addresses linear system identification for aircraft using artificial neural networks. The output of a linear aircraft system consists of linear combinations of state and control inputs. Determining linear models for aircraft is historically a very time consuming process for obtaining accurate results. Methods like Observer/Kalman Filter Identification are often used to determine these linear models by analyzing flight data under specific flight conditions and requiring extended labor on the part of the user to determine the correct model. In this paper, a new method of system identification is proposed that uses artificial neural networks specially designed for determining the linear model of an aircraft. This method, called Artificial Neural Network System Identification, has the advantages of being straightforward with low computational burden. Results presented in this paper demonstrate that it is capable of accurately determining a linear model in under 8 seconds of CPU time, and comparisons to Observer/Kalman Filter Identification show that it also has the potential to be the more accurate model. 2013 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.
name of conference
51st AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition
