Identify aerodynamic derivatives of the airplane attitude channel using a spiking neural network

Abstract

Verbatim from Embry-Riddle Scholarly Commons. Not paraphrased, not summarized.

The paper proposes a method for identifying aerodynamic coefficient derivatives of aircraft attitude channel using spiking neural network (SNN) and Gauss-Newton algorithm based on data obtained from actual flights. Using SNN combination with Gauss-Newton iterative calculation algorithm allows the identification of aerodynamic coefficient derivatives in a nonlinear model for aerodynamic parameters with higher accuracy and faster calculation time. The paper proposes an algorithm to train the SNN multi-layer network by Normalized Spiking Error Back Propagation (NSEBP), in which, in the forward propagation period, the time of output spikes is calculating by solving quadratic equations instead of detection by traditional methods. The phase of propagation of errors backward uses the step-by-step calculation instead of the conventional gradient calculation method. The identification results are compared with the results when using the RBN network to prove the algorithm efficiency

Authors

• VInh, Nguyen Quang Embry-Riddle Aeronautical University
• Duc Thanh, Nguyen Embry-Riddle Aeronautical University
• Minh Dac, Hoang Embry-Riddle Aeronautical University
• Dang Khoa, Truong Embry-Riddle Aeronautical University

Keywords

• Aerodynamic identification
• Nonlinear model
• Flying vehicle
• Aerodynamics and Fluid Mechanics
• Other Aerospace Engineering
• Systems Engineering and Multidisciplinary Design Optimization