This paper presents a method to control and identify the servo pneumatic system using a mixed reality environment. A mathematical model is presented to study the system dynamics and nonlinear effects of the servo pneumatic system. The auto-regressive moving-average (ARMA) model-based recursive least squares (RLS) algorithm is utilized to identify the transfer function of the servo pneumatic system in a real-time environment. The identification of the servo pneumatic system can be carried out effectively and efficiently using the proposed ARMA model. Furthermore, the high precision to identify the system with minimum error, and reducing the time in adjusting the parameters of the control unit. The discrete transfer function of the servo pneumatic system is identified in real-time from the inputs and outputs data of the system. The identification results showed that the fourth-order system model achieved the minimum square error with one-step prediction. The experimental results showed the accuracy and effectiveness of the proposed method.