In this thesis study, modeling and control of the in-house developed twin rotor systems are aimed. Firstly, two input-output models are obtained by using experimentally collected data. One model is obtained as transfer functions in the Laplace domain while the other is a neural network based model. Nextly, lag and lead type compensators are designed and then experimentally verified on the twin rotor system. Specifically, first, lag and lag-lag compensators are designed to obtain a reduced steady state error when compared with proportional controllers. Secondly, lead compensation is discussed to obtain a reduced overshoot. Finally, to make use of their favorable properties at the same time, lag-lead compensators are designed. All the compensators are applied to the twin rotor system in our laboratory.