Because of its many advantages, the minimally invasive surgery (MIS) becomes more and more important as a medical technique. During MIS, the operation area has to be flushed with a rinsing fluid for a suitable expansion and to maintain visibility. This is done with the help of a special medical device called double roller pump (DRP). By controlling the pump’s speed for the inflow and outflow of fluid, the pressure in the operation area is changed respectively. It is possible to design a controller for a stable pressure by using directly measured pressure in the surgical area. However, in the case of a real-life operation, the pressure in the part of the body cannot be measured directly because of the safety of the patient. Therefore, this paper presents a solution to reconstruct the pressure in the operation area by using the available sensor data at the double roller pump for the application of a knee arthroscopy, as an example of MIS. A Luenberger observer and a Kalman filter were implemented on simulation as well as on a real-life simulator. The estimated results illustrate that the Kalman filter works better than the Luenberger in terms of fast-tracking and noise cancelling.