Real-time trajectory tracking of a 2-DoF helicopter using discrete sliding modes with neural block control

This paper addresses a solution of the trajectory tracking problem for a two degree of freedom (2-DoF) helicopter considering plant parameter variations. The helicopter is identified using a discrete-time Recurrent High-Order Neural Network trained on-line with an extended Kalman filter algorithm, and it is later transformed to the block control form (BCF) to design a sliding surface, it which depend on the tracking errors. Finally, a non-switching discrete-time sliding mode control is applied to ensure that the sliding surface be attractive. The effectiveness of the proposed controller is demonstrated by a real-time implementation.