Real-time neural backstepping control for a helicopter prototype

Abstract

This paper presents a discrete-time backstepping controller based on a neural model for a Quanser 2-Degree Of Freedom (DOF) helicopter. The proposed controller is used to track the pitch and yaw position references independently. This controller is based on a Recurrent High Order Neural Network (RHONN) trained with an Extended Kalman Filter (EKF). The RHONN works as an identifier to obtain an adequate Quanser 2-DOF helicopter mathematic model, which is robust in presence of disturbances and parameter variations. To examine the robustness of the proposed controller, simulations using Matlab/Simulinkand real-time implementation are presented.