Abstarct: One of the most important astronomical instruments are large telescopes, whose primary mirror surface accuracy is about 2 nm. Factors such as rotation of the altitude axis, rapid change of azimuth axis, wind blowing, etc. can cause great distortions in its surface. In order to compensate for the mentioned factors and maintain the accuracy of the mirror surface, actuators are installed under it to compensate the added forces. These actuators must be able to produce force in a wide range and with an accuracy of about a few tenths of newtons in the shortest possible time. In recent years, due to the many advantages that pneumatic actuators have, they have received more attention. But the main problem of these actuator is their non-linearity and difficult control. The aim of this thesis is to design and build a suitable controller for a model of actuators that can meet the required requirement. According to the nonlinear behavior and the results of practical implementations, conventional proportional-integral-derivative linear controllers do not show proper response. On the other hand, due to hardware limitations, very complex structures cannot be used for the controller. baxsed on the study, the use of fractional order controllers is suggested. Since the system model is not available and its identification, in addition to existing limitations, leads to complex models and high uncertainties, an independent method of iterative feedback tuning model is used to adjust the controller parameters. One of the most important problems of this method, which operates on the basis of gradient descent, is falling in the local minimum points, which is used to solve this problem by combining it with the genetic algorithm. After simulating different systems and ensuring the correctness of the operation, the proposed method is implemented on an actautor and the results confirm its efficiency.