Self-organizing intelligent system and its application for hard disk drive control

An intelligent system with self-organizing ability is discussed in this paper. The intelligent system is based on computational models of fuzzy inference, neural-processing, and machine learning. The concept of pseudo errors is utilized in the study. Clusters are generated from the pseudo-error information, and they are corresponding to fuzzy rules of the self-organizing neuro-fuzzy system (SO-NFS). In this paper, a learning scheme is proposed for hard disk drive (HDD) motion control. The SO-NFS is used as a controller to the HDD servo system for both the track-seeking and the track-following motion control. A reference speed-displacement curve is used to guide the control process. Good performance of the proposed approach is observed.