<p>This study examines methods to address the challenges of unknown nonlinearity and parameter uncertainty in the control process of single-rod hydraulic cylinders. First, the unknown nonlinearities contained in the mathematical model of the hydraulic system are collectively modeled to obtain a multi-source disturbance term. Then, the multi-source disturbance is reconstructed into static and dynamic terms. Next, a parameter adaptation mechanism is incorporated into the controller design to adjust the uncertain parameters in the system. A fuzzy neural network (FNN) based on meta-cognitive learning is employed to approximate the static disturbance term in the system. Moreover, an extended state observer (ESO) is used to estimate the unknown states and dynamic disturbance terms in the system. In addition, the ESO incorporates the structural errors of the neural network, enabling accurate estimation of unknown states and dynamic disturbance terms of the system using only position feedback information. The structure of the meta-cognitive fuzzy neural network (McFNN) can be adaptively adjusted according to the system’s dynamic characteristics, reducing computational complexity while ensuring modeling accuracy. The ESO and McFNN are embedded within the adaptive backstepping control framework to achieve joint compensation control of multi-source disturbances in the system, suppressing the impact of multi-source disturbances in a nonlinear system on the control performance. Finally, the proposed method is verified in terms of control precision and disturbance resistance through stability analysis and experimental comparison under various working conditions.</p>