This paper presents a mathematical model of the Kalman Extended Filter (EKF) for estimating dynamic states in an electro-hydraulic system. The main goal is to increase the accuracy and stability of estimates in the presence of noise and nonlinearities in the dynamics of the system. EKF combines information from hydraulic motor shaft angular velocity measurements and the pressure difference between its inlet and outlet to provide reliable estimates of basic dynamic system parameters, including the current angular velocity and the moment created by the external load. Simulation studies demonstrate the advantages of the EKF compared to a system without a filter, emphasizing the importance of tuning the covariance matrices Q and R to achieve a balance between response speed and filter stability. The developed model shows high applicability for real electro-hydraulic systems, while offering the possibility of adaptation to other nonlinear dynamic systems.