In this study, a state-of-the-art methodology for controlling an electro-hydraulic system is proposed. The aim is to achieve superior position tracking performance comparable to that of an electro-hydraulic servo valve system. To achieve this, a suite of linear and nonlinear control techniques—including PID, LQR, sliding mode, model predictive control (MPC), and neural network MPC controllers—are designed and tested based on system dynamics approximation. The controllers are optimized to effectively address the challenges posed by various loads, uncertainties, nonlinearities, internal leakage, chattering, and overshooting in the electro-hydraulic system. The proposed approach is both practical and effective, as demonstrated by simulation and experimental results. Comparative analysis reveals that the neural network MPC controller exhibits exceptional tracking performance and stability, with a smooth response and quick settling time.