A Vision Based Rapid Alignment System for Short Pitch Movement Using Dual Stage Servo System

碩士 === 國立高雄第一科技大學 === 系統資訊與控制研究所 === 96 === This research is aiming to combine an ordinary servo platform with a piezo-actuated positioning platform, in which a machine vision system is required for position feedback. However, utilizing machine vision as a position feedback will introduce a signific...

Full description

Bibliographic Details
Main Authors: Wen-Shian Wang, 王文憲
Other Authors: Yuan-Chen Yu
Format: Others
Language:zh-TW
Published: 2008
Online Access:http://ndltd.ncl.edu.tw/handle/77549742598148615493
Description
Summary:碩士 === 國立高雄第一科技大學 === 系統資訊與控制研究所 === 96 === This research is aiming to combine an ordinary servo platform with a piezo-actuated positioning platform, in which a machine vision system is required for position feedback. However, utilizing machine vision as a position feedback will introduce a significant time delay into the servo control loop. Such a time delay in the control loop will significantly decrease the stability and bandwidth of the control system. Therefore, this research is mainly to analyze the effect of such vision time delay to the performance of the dual stage alignment servo system and trying to design a compensation algorithm for it. Traditional compensation approaches were used firstly to see how good the traditional approach can reach. Finally, a new approach, called One Shot Alignment on Fly (OSAOF), was proposed in this research. Using this approach, image was taken only once, and the vision time delay was got rid of from the control loop. The alignment speed was increased significantly and the alignment can be performed even while the long stroke platform is still in motion. Simulation and experimentation have shown that the proposed approach can achieve much better alignment performance then the traditional approaches.