Abstract High speed and high accuracy mechanical systems pose an increasing demand on the control systems to eliminate the influence of disturbances like cogging and friction. One way to incorporate additional disturbance rejection for linear systems is to extend the controller with a disturbance observer. By virtue of the separation property, the disturbance observer and the controller can be designed separately. In this paper, an overview is given of choices to be made to design a disturbance observer, and the consequences of every step are discussed. It follows that, despite the separation property, knowledge about the feedback controller is needed to design the disturbance observer (and vice versa), in order to guarantee robust stability. Existing disturbance observer design techniques are extended to account for model uncertainty. Experimental results on a linear permanent magnet motor demonstrate the effectiveness and robustness of the disturbance observer design method.
[1]
Masayoshi Tomizuka,et al.
Robust motion controller design for high-accuracy positioning systems
,
1996,
IEEE Trans. Ind. Electron..
[2]
Yoichi Hori,et al.
Robust speed control of DC servomotors using modern two degrees-of-freedom controller design
,
1991
.
[3]
Theodorus J.A. de Vries,et al.
Linear motor motion control using a learning feedworward controller
,
1997
.
[4]
Masayoshi Tomizuka,et al.
Passivity-based versus disturbance observer based robot control : Equivalence and stability
,
1999
.
[5]
Hui Zhang,et al.
H∞ control versus disturbance-observer-based control
,
1998,
IEEE Trans. Ind. Electron..
[6]
Carl J. Kempf,et al.
Disturbance observer and feedforward design for a high-speed direct-drive positioning table
,
1999,
IEEE Trans. Control. Syst. Technol..