This paper presents the study, manipulation and control of a two degrees of freedom system, known as planar pendulum, using reaction wheels to achieve a certain position, the basic concept of this system being the dynamics of an inverted pendulum. The classic PID controller was proposed as well as a Fuzzy control. The mathematical model was generated to be used in MatLab numerical simulations of these controls, using the Simulink tool which allows a self-adjustment of the controller of the system. The PID control was built according to the self-tuning of Simulink and also using the Ziegler Nichols method. The differences between both are discussed. Fuzzy control was designed, creating the bank of rules and looking for a control strategy more suitable for the system. Friction in joints, and drives were taken in account in the mathematical modeling. Finally a real model was built using a data acquisition board in conjunction with LabView for control and a microcontroller board called Arduino, as well as an encoder for the acquisition of angles. Measurement errors and unbalance of the system are problems that could not be completely eliminated, but were kept to a minimum. Numerical and experimental results were compared for each control and for each assembly and their differences were discussed.