Due to not only having strong nonlinear inter-couplings in its model but also being an open-loop unstable system, control of a 2-degree of freedom (DOF) helicopter is a challenging task. This chapter deals with the decentralized control of the Quanser 2-DOF helicopter system by designing an interval type-2 fuzzy neural network for the control of the pitch and yaw angles by using a sliding mode control theory-based training algorithm. The proposed control method is known as feedback error learning in which an intelligent controller, a type-2 fuzzy neural network in this case, works in parallel with a conventional PD controller. In the proposed scheme, on one hand, the conventional PD controller is responsible to maintain the stability of the system until the intelligent controller takes the responsibility of controlling the system. On the other hand, the intelligent controller learns the system dynamics online with a sliding mode control-theory based learning algorithm. The simulation results show that without having neither a priori knowledge about the mathematical model of the system nor its parameters, the proposed control algorithm is able to track the reference signals for both yaw and pitch angles without giving a steady state error. In addition, the simulation results show the superiority of the proposed control scheme over its type-1 counterpart in the presence of noise in the system. In addition to its robustness, the sliding mode control theory-based learning algorithm has additional advantages such as having no matrix manipulations or partial derivatives which makes the overall training and control algorithm computationally simple and fast when compared to other methods, e.g. gradient-descent based methods.