Designing Dynamic Positioning System Based on H∞ Robust Recurrent Cerebellar Model Articulation Controller

This research proposes a H
^∞
robust recurrent cerebellar model articulation controller (RRCMACS) for dynamic positioning system (DPS). In this system, the fast learning, generalization capability of the CMAC and the recurrent technique of neural network incorporated to mimic an ideal force controller. Furthermore, the H
^∞
robust controller efficiently vanishes the effects of external disturbances such as wave, wind and current acting on the ship to achieve the robustness of the system. The parameters of the controller are learned and adjusted in the sense of the Lyapunov-like Lemma theory. Therefore the stability of the system is guaranteed. The simulation results for the dynamic positioning system are given to prove the effectiveness of the proposed control system for the DPS and model-free nonlinear systems.