Search

This paper considers the set-point control problem of an uncertain six-degree-of-freedom precision positioning system, where only position measurements are available. As a solution, we present an iterative output feedback control algorithm whose learning mechanism is based on the concept of contraction mapping. Besides stability and convergence, the presented analysis includes practical aspects associated with the algorithm implementation. In particular, it is shown that: 1) the system response due to the iterative process converges to an arbitrarily small neighborhood of the desired equilibrium point, and 2) the steady-state response after each iteration is bounded inside the system traveling range. The presented control algorithm is demonstrated experimentally.

This paper presents a control scheme for a linear time invariant (LTI) system with uncertainty and disturbance, which are not assumed to be matched with the control input. The presented control scheme consists of a stabilizing feedback controller and a newly developed feed forward mechanism, which counteracts the uncertainty and disturbance through the reference path. In contrast to the related control methods, such as time delay control (TDC) and uncertainty and disturbance estimation (UDE) based control, the scheme presented in this paper does not impose any structural constraints on the controlled plant. The presented control scheme was verified experimentally using a flexible mechanical testbed with a poorly damped resonance.

AMB (Active Magnetic Bearing) technology based multi-degree-of-freedom positioning systems is a very promising solution to many precise motion problems. In some applications, such as the photolithography process, AMB based positioning system can be used to carry tools of unknown geometry and mass. To avoid the positioning stage tracking performance degradation due to unknown payload, adaptive control methods should be used. Although the subject of adaptive control algorithms for the AMB based rotating machinery was widely studied in the past, due to some differences, the available results can not be applied in a straight-forward manner to the precise AMB based positioning systems. This paper contributes to the field of precise AMB based positioning stages by presenting design and implementation of an adaptive control algorithm for a 1DOF (degree-of-freedom) AMB based suspension system. In this study we assume that the payload and electromagnetic coefficient values of the plant are unknown and only the output measurements are available as a feedback. In addition, the control currents computed by the presented algorithm are free of a bias term. The control law in this papers solves the trajectory tracking problem, and its effectiveness was verified experimentally.

This paper presents the concept and the practical implementation of a switch-mode power driver for the use in Active Magnetic Bearings (AMB). The switch-mode driver consists of a full-bridge capable of operating in 2-level, 3- level, or phase-shifted PWM mode. An optimal digital current- programmed mode controller lays out the foundations to a proximate-optimal, robust, large-signal dynamic response of the system under extreme cases of loading and parameters variations. In particular, the developed power driver is capable of following the desired reference for large inductance variations (due to displacement). The power stage combines a newly developed, efficient, passive-based current sensor that is capable of sensing bi-directional and near-zero currents, while being robust to exter- nal disturbances and maintaining simple and efficient structure. Performance evaluation of the power driver has been verified experimentally on a closed-loop design for a single-axis AMB platform using a dual channel prototype, controlled using a single 16-bit, low-cost microcontroller. I. INTRODUCTION

This paper presents modelling and control of a novel positioning stage based on Active Magnetic Bearings (AMB) technology. The stage includes a single moving part levitated and controlled with 6 degrees of freedom by 6 couples of reluctance electromagnetic actuators. The position control of the stage is achieved by simultaneous stabilization of each one of its 6 actuators. The proposed control law is an output feedback nonlinear controller combined with a velocity observer. This observer is based on the concept of approximate differentiation. The main feature of the presented positioning stage is that all 6-DOF are controlled by electromagnetic forces generated by AMB actuators. No other means, such as permanent magnets or mechanical support, are present in order to counteract the payload's weight against gravity.