A Vision-Based Automated Manipulation System for the Pick-Up of Carbon Nanotubes

The ability to pick up a single carbon nanotube (CNT) from a bundle of CNTs is of great importance for nanodevice fabrication. In this study, we propose a nanorobotic manipulation system allowing automated pick-up of CNTs based on visual feedback. We used histogram thresholding for automatic binarization, and it clearly distinguished CNTs from the substrate and other impurities under various image brightnesses and contrasts. Furthermore, the CNT tip was successfully extracted by making use of the geometrical characteristics of the CNT. We designed a segment detection method to separate the CNT and atomic force microscope cantilever during overlapping. The contact detection between them was identified by evaluating the linearity of the fitted CNT curve. We also further analyzed the specific properties of point contact and linear contact, significantly improving the success rate of pick-up. Finally, the experimental results show that our method is highly promising for realistic fabrication of nanodevices.