In this paper, a sliding mode control with integral action for a Twin Rotor Multiple Input Multiple Output (MIMO) System is presented. The nonlinear controller is based on a estimated model of the Twin Rotor System obtained by using a Recursive Least Squares (RLS) identification algorithm. The performance of the proposed adaptive control method is evaluated over a simulated and a real model. It is noticeable, that the proposed method outperforms, in terms of tracking error and settling time, the classical MIMO controllers under internal and external disturbances. As a result, it can be seen that the sliding mode control algorithm with integral action based on a RLS identification method is suitable for real applications where the model parameters are not well defined.