Title: Aircraft Parameter Estimation Using Neural Network Based Algorithm
Abstract: Aircraft Parameter estimation is probably the most outstanding and illustrated example of the system identification methodology. In the past the most widely used parameter estimation methods have been Equation error method, Output error method, Maximum likelihood method and Filter error method. In this paper an algorithm based on neural modeling and Gauss-Newton optimization is proposed to estimate aerodynamic parameters from flight data. The proposed method bypasses the requirement of solving the equations of motion to predict motion variables. The proposed algorithm uses a black box approach to build the flight dynamic model of an aircraft using measured flight variables. The algorithm was initially validated on flight data generated using HANSA-3 aircraft at Indian Institute of Technology Kanpur. Similar validation was carried out with lateral-directional flight data of ATTAS aircraft (supplied by DLR Germany). The estimated values were compared with estimates obtained using Filter error method and Least square method. The applicability of the proposed algorithm in handling flight data with atmospheric turbulence was also investigated by applying it on real flight data of test aircraft, HFB-320. The work concludes by presenting a comparison among the estimates obtained by applying several methods on real flight data of three different test aircraft.
Publication Year: 2009
Publication Date: 2009-06-14
Language: en
Type: article
Indexed In: ['crossref']
Access and Citation
Cited By Count: 24
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot