Active Magnetic Bearing system (AMB) is a mechatronic device which is used to levitate rotating parts of a machine without contact to the stationary part of the machine. It has nonlinear characteristic, multi-input multi output and unbounded input output systems. The non-linearity properties should be coming from the relation between force, current and position of the rotor. Due to those characteristics, it is challenging for the system stability. In order to maintain a stable and good performance; controllers must be designed as part of the system. The mathematical model was developed and it shown that the AMB system considered as a 2X2 MIMO system. In order to reduce the complexity of the system, the nonlinear model is linearized by appropriate linearization method. The advantage and application area of active magnetic bearings has been defined. The simulation results were carried out by Matlab/Simulink version 2018a. In this thesis, Neuro fuzzy sliding mode controller has been designed for control the position of active magnetic bearing system. Matlab Simulink models of AMB with SMC, FSMC and NFSMC are developed to carry out simulation studies. The comparison results show that with SMC; the position of the rotors is following the reference input but it affected by chattering. It is further observed that the FSMC are reduced the chattering, and also NFSMC controllers are eliminate automatically the chattering problem and it is more robust and smoothly operate. The stability of the system has been analyzed by using Lyapunov theorem. The simulation result show that the overall system output performance can be improved using the proposed controller, the NFSMC resulted no overshoot, small settling time(0.5sec) and less rise time (0.10 sec) as compared to 0.392%, 0.75sec and 0.133sec for FSMC controller, and 9.047%, settling time 7.5sec and rise time 0.17sec for existing SMC. Key words: Active magnetic bearing, neural network, fuzzy logic, sliding mode controller, Matlab/Simulink