Resist based lithographical techniques are widely applied for graphene processing. These resists can leave residues leading to parasitic effects that deteriorate the desired properties of graphene. This paper presents an experimental setup tailored for resist-free robotic processing of graphene with in-situ vision based control. A robust graphene detection and classification approach is presented applying multiple image processing operations of the visual feedback provided by a high-resolution light microscope. Detected graphene flakes can be modified using scanning probe based lithographical processes, such as mechanical and bias-assisted approaches, that are directly linked to the in-situ optical images. The results of this process are discussed with respect to further application scenarios.

基于抗蚀剂的光刻技术已广泛应用于石墨烯加工。这些抗蚀剂会留下残留物，从而导致寄生效应，从而降低石墨烯的所需性能。本文介绍了一种专为基于原位视觉控制的石墨烯无抗蚀剂机器人加工而量身定制的实验装置。提出了一种鲁棒的石墨烯检测和分类方法，该方法应用了高分辨率光学显微镜提供的视觉反馈的多个图像处理操作。可以使用基于扫描探针的光刻工艺（例如机械方法和偏置辅助方法）对检测到的石墨烯薄片进行修饰，这些方法直接与原位光学图像相连。将针对其他应用场景讨论此过程的结果。