Modeling nonlinear features of V tail aircraft using MNN

The nonlinear stability and control modeling involves a very popular high performance general aviation aircraft that uses a "V" tail assembly instead of the traditional inverted "T" tail. The nonlinear response features of this aircraft result from a well developed Dutch roll mode and are caused by a dynamic stall phenomenon that occurs on the V tail during the maneuver. A new approach using multilayered neural network (MNN) for modeling the nonlinear features of this aircraft is suggested here. Both the conventional backpropagation (BP) and the extended Kalman filter (EKF)-based learning algorithm are used for training the neural network. Simulation results that confirm the efficacy of the method are given. Further, performance comparison of the EKF-based and the conventional BP algorithm is made to highlight the effectiveness of the EKF-based learning algorithm.<>