Adaptive Tuning and Control of a Hard Disk Drive

This paper deals with the implementation of adaptive tuning and control on a computer hard disk drive (HDD). A model-based adaptive loop is added to a LTI stabilizing controller to minimize the position control error of the HDD head. This adaptive loop may be used in either of two modes: control tuning (CT) or real-time fully adaptive control (RTFAC). The former (CT) consists of adapting a FIR filter to compensate for unknown disturbance dynamics using the inverse QR-RLS algorithm. Once the filter has converged it is wrapped into the LTI controller and a new tuning process is performed, iteratively improving the control performance. The second method (RTFAC) consists of adapting indefinitely a FIR filter to compensate for unknown disturbances. This method is implemented using an unwindowed lattice RLS filter, which makes the scheme converge substantially faster than when using other adaptive algorithms. This pertains to the HDD data seek time. Another innovation in this paper is the introduction of a frequency-weighting filter in the RTFAC scheme, which accounts for robust stability. Experimental results are presented for both adaptive tuning and control methods, programmed on a commercial hard disk drive.