標題: Sliding Mode Control Using Virtual Eigenvalue Method for Compact Optical Image Stabilization Actuators
作者: Yu, Hsing-Cheng
Liu, T. S.
機械工程學系
Department of Mechanical Engineering
關鍵字: Image stabilization;linear actuator;sliding mode control
公開日期: 1-Nov-2008
摘要: The latest research in optical image stabilization actuators (OISA) focuses on improving the image quality of high resolution cameras in a photographing optical system (POS). The focal image blur is always caused of external and environmental jitter when pictures of an image sensor are acquired by photographers. Several works concerned for OISA are emphasized mainly OISA design methods; consequently, they are devoted to minimizing dimensions and maximizing driving magnetic force with low power consumption. Although two proportional-integral-derivative controllers and lead-lag compensators have been proposed to position dual axes of OISA in a POS, they lack robustness to compensate nonuniform friction, system uncertainty, and external disturbance. Therefore, this work aims at the sliding mode control (SMC) using virtual eigenvalue method to achieve fast time response and accurate position despite vibration from external jitter and friction of a compact OISA. The sliding surface can be reached in finite time to match disturbance. Accordingly, the SMC algorithm works well despite system uncertainty and external disturbance, and then the time response of simulation and experiment are matched. In consequence, the settling time of dual axes are converged within 0.2 s when the dual axes of movable platform of a compact OISA move at 0.5 mm stroke in a POS.
URI: http://dx.doi.org/10.1109/TMAG.2008.2002787
http://hdl.handle.net/11536/28765
ISSN: 0018-9464
DOI: 10.1109/TMAG.2008.2002787
期刊: IEEE TRANSACTIONS ON MAGNETICS
Volume: 44
Issue: 11
起始頁: 4074
結束頁: 4077
Appears in Collections:Conferences Paper


Files in This Item:

  1. 000262221300180.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.