A prediction-based optimal gain selection in RISE feedback control for hard disk drive

Manel Taktak-Meziou (1) , Ahmed Chemori (2) , Jawhar Ghommam (1) , Nabil Derbel (1)

1 ENIS - École Nationale d'Ingénieurs de Sfax | National School of Engineers of Sfax
2 DEXTER - Robotique médicale et mécanismes parallèles

This paper presents a prediction-based optimal gain selection in Robust Integral Sign of the Error (RISE) based Neural Network (NN) approach. Previous research has shown that combining a feedforward term with a feedback control element yields an asymptotically stable closed-loop system. The proposed approach adds a prediction-based optimal technique which minimizes a quadratic performance index to calculate an optimal feedback gain. The resulting novel controller, called PRISE -NN, is applied for a track following problem of a Hard-Disc-Drive servo-system. Simulation studies are used to show the efficiency of the proposed control scheme and its robustness against external disturbances and parametric uncertainties in the system. The authors believe that the proposed control solution combining RISE with a predictive control approach has never been conducted before.

Domaines: Automatique / Robotique

Manel Taktak-Meziou, Ahmed Chemori, Jawhar Ghommam, Nabil Derbel. A prediction-based optimal gain selection in RISE feedback control for hard disk drive. CCA: Conference on Control Applications, Oct 2014, Juan Les Antibes, France. pp.2114-2119, ⟨10.1109/CCA.2014.6981615⟩. ⟨lirmm-01723921⟩