Landmark based sensing and positioning in robotic nano manipulation

AFM based nanomanipulation is still hindered by large spatial uncertainties encountered in tip positioning caused by PZT nonlinearity and thermal drift. In this paper, a landmark based positioning method is proposed to solve these problems. Its pivotal idea is that the tip position is described in a feature based landmark coordinate instead of length based Descartes coordinate. During manipulation, the landmarks selected from surface features, which is sensed through the new developed local scan method, acts as coordinate reference for tip localization. In this way, the positioning error aroused from PZT nonlinearity and thermal drift can be effectively suppressed. The details about landmark selecting, sensing, identifying will be discussed. Experiments of nanoparticle manipulation are presented to demonstrate the efficiency and effectiveness of the proposed method.