Dynamic Surface-Based Prescribed-Time Robust Adaptive Control of Hydraulic Manipulators

Hydraulic manipulators suffer from joint coupling, strong disturbances, and highly nonlinear dynamics, all of which significantly degrade motion control performance. This paper proposes a dynamic surface-based prescribed-time robust adaptive controller. A complete manipulator dynamic model incorporating both mechanical and hydraulic dynamics is first derived. Within a backstepping framework, a nonlinear composite filter is employed to construct dynamic surfaces for the virtual control laws, enabling fast error decay while maintaining smooth filtering. Then, adaptive laws are then developed to address parameter uncertainties in both the mechanical and hydraulic systems, while the nonlinear robust terms are incorporated to rapidly attenuate modeling errors and unknown disturbances. Furthermore, a prescribed-time performance function is integrated into the dynamic surface design to guarantee that the tracking error converges within predefined bounds at a specified time, thereby improving the transient behavior of the closed-loop system. Lyapunov-based analysis demonstrates that the proposed controller achieves asymptotic tracking. Simulation results on a 3-degrees of freedom (DOF) hydraulic manipulator illustrate the effectiveness of the proposed approach.