Design and implementation of a vision system for microassembly workstation
Doğan, Eray (2006) Design and implementation of a vision system for microassembly workstation. [Thesis]
Rapid development of micro/nano technologies and the evolvement of biotechnology have led to the research of assembling micro components into complex microsystems and manipulation of cells, genes or similar biological components. In order to develop advanced inspection/handling systems and methods for manipulation and assembly of micro products and micro components, robust micromanipulation and microassembly strategies can be implemented on a high-speed, repetitive, reliable, reconfigurable, robust and open-architecture microassembly workstation. Due to high accuracy requirements and specific mechanical and physical laws which govern the microscale world, micromanipulation and microassembly tasks require robust control strategies based on real-time sensory feedback. Vision as a passive sensor can yield high resolutions of micro objects and micro scenes along with a stereoscopic optical microscope. Visual data contains useful information for micromanipulation and microassembly tasks, and can be processed using various image processing and computer vision algorithms. In this thesis, the initial work on the design and implementation of a vision system for microassembly workstation is introduced. Both software and hardware issues are considered. Emphasis is put on the implementation of computer vision algorithms and vision based control techniques which help to build strong basis for the vision part of the microassembly workstation. The main goal of designing such a vision system is to perform automated micromanipulation and microassembly tasks for a variety of applications. Experiments with some teleoperated and semiautomated tasks, which aim to manipulate micro particles manually or automatically by microgripper or probe as manipulation tools, show quite promising results.
Repository Staff Only: item control page