Modelling of Aircraft’s Dynamics Using Least Square Support Vector Machine Regression
The system identification is a broad area of research in various fields of engineering. Among them, our concern is to identify the aircraft dynamics by means of the measured motion and control variables using a new approach which is based on the support vector machine (SVM) regression. Due to the computational complexity of SVM, it is suggested to adopt the advanced version of SVM i.e. least square support vector machine (LSSVM) to be used for system identification. LSSVM regression is a network-based approach which requires a user defined kernel function and a set of input-output data for its training before the prediction phase like a neural-network (NN) based procedure. In this paper, LSSVM regression has been used to identify the non-linear dynamics of aircraft using real flight data.
KeywordsSystem identification LSSVM regression Kernel function
- 2.Hamel, P.G., Jategaonkar, R.V.: The evolution of flight vehicle system identification. AGARD, DLR Germany, 8–10 May 1995Google Scholar
- 8.Haykins, S.: Neural Networks: A Comprehensive Foundation. McMaster University, Macmillan College Publishing Company, New York (1994)Google Scholar
- 9.Raisinghani, S.C., Ghosh, A.K., Kalra, P.K.: Two new techniques for parameter estimation using neural networks. Aeronaut. J. 102(1011), 25–29 (1998). UKGoogle Scholar