Browse Prior Art Database

Robust Continuous Repetitive Disturbance and Rejection

IP.com Disclosure Number: IPCOM000117853D
Original Publication Date: 1996-Jun-01
Included in the Prior Art Database: 2005-Mar-31
Document File: 4 page(s) / 128K

IBM

Related People

Cheung, W: AUTHOR [+2]

Abstract

Disclosed is a method for robust and continuous estimation of track runout in disk drive systems and the precise positioning of the actuator over a data track for read/write operations.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Robust Continuous Repetitive Disturbance and Rejection

Disclosed is a method for robust and continuous estimation of
track runout in disk drive systems and the precise positioning of the
actuator over a data track for read/write operations.

The method embodied in this disclosure is to incorporate a
disturbance model with unknown states and to make use of an estimator
to robustly and continuously estimate the current states of the
runout disturbance.  The estimated states of the disturbance can then
be used to generate a correction control input to the actuator.  For
repetitive disturbance e, this form of the repetitive frequencies
(including DC) and guarantees robust asymptotic tracking and/or
disturbance rejection.  The structure of the estimator remains the
same during seeking and thus the correction control input need not be
changed when the controller mode changes from tracking to seeking or
vice versa.

A sinusoidal signal, r((wr))(k), can be generated using the following
model: (refer to original document for mathematical formula).
where w(r) is the angular frequency, rl(wr)o and r2(wr)o are the
unknown initial states of the sinusoidal signal and T(s)  is the
sampling time.

Similarly, a DC signal, r(o)(k), is generated by: (refer to original
document for mathematical formula).

A repetitive signal, w(k), with period T, and T over <T>(s) = N has
m = lb <N over 2>rb repetitive frequencies (including DC), where lb x
rb denotes the largest integer le x.  Let w(k)5=r(0)(k)+r(w1)(k)+...
+r(wm)(k)=w(k-N) w(k) can be written in state space form as follows:
(refer to original document for mathematical formula).
where: (refer to original document for mathematical formula).
A system (A(c), B(c), C(c)) with disturbance w(k)at the input can
be written in state space form as: (refer to original document for
mathematical formula).

Using the disturbance model in Eqn (3), Eqn (5) can be rewritten as:
(refer to original document for mathematical formula).

An estimator can be designed using Eqn (5) that assures: (refer to
original document for mathematical formula).
where (hat) are estimates of the states of the system and
disturbance generator in Eqn (7).  Thus, the disk repetitive runout
can be followed by letting the control input, u(k), to the plant (or
actuator) be: (refer to original document for mathematical formula).
where the state feedback gain K can be obtained through usual state
feedba...