2020/04/06 by Maik Pfefferkorn, Michael Maiworm, Pfefferkorn, Maik +7
Biochemistry, Genetics and Molecular Biology · #Advanced Fluorescence Microscopy Techniques #FOS: Electrical engineering #FOS: Physical sciences #Instrumentation and Detectors (physics.ins-det) #Systems and Control (eess.SY) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2004.02488
openalex publication_date 2020/04/06 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
Elucidating electrostatic surface potentials contributes to a deeper\nunderstanding of the nature of matter and its physicochemical properties, which\nis the basis for a wide field of applications. Scanning quantum dot microscopy,\na recently developed technique allows to measure such potentials with atomic\nresolution. For an efficient deployment in scientific practice, however, it is\nessential to speed up the scanning process. To this end we employ a\ntwo-degree-of-freedom control paradigm, in which a Gaussian process is used as\nthe feedforward part. We present a tailored online learning scheme of the\nGaussian process, adapted to scanning quantum dot microscopy, that includes\nhyperparameter optimization during operation to enable fast and precise\nscanning of arbitrary surface structures. For the potential application in\npractice, the accompanying computational cost is reduced evaluating different\nsparse approximation approaches. The fully independent training conditional\napproximation, used on a reduced set of active training data, is found to be\nthe most promising approach.\n