Regensburg 2022 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
O: Fachverband Oberflächenphysik
O 6: Nanostructures at Surfaces 1
O 6.1: Talk
Monday, September 5, 2022, 10:30–10:45, S052
Autonomous molecular manipulation of nanocars based on reinforcement learning — •Bernhard R. Ramsauer1, Grant J. Simpson2, Johannes Cartus1, Andreas Jeindl1, Leonhard Grill2, and Oliver T. Hofmann1 — 1Institute of Solid State Physics, NAWI Graz, Graz University of Technology, Petersgasse 16, 8010 Graz, Austria — 2Institute of Chemistry, NAWI Graz, University Graz, Heinrichstraße 28/IV, 8010 Graz, Austria
At the world's first race of nanocars at the CEMES-CNRS, in France, participants had to direct a nanocar across a specific *racetrack* [1]. In order to control their nanocar, they have to manipulate it via an STM-tip, without being in direct contact with the nanocar. The physics that govern the molecule*s movement and rotation is complex and involves the interaction between the tip and the molecule as well as the molecule and the substrate [2]. Thus, it requires time and effort for humans to be able to maneuver a molecule with a reasonable success rate. However, predicting the outcome of a performed action is unintuitive and often hard to predict for humans. Therefore, we developed an artificial intelligence (AI) based on reinforcement learning (RL) and show how it can be implemented to manipulate single molecules. The AI utilizes an off-policy learning algorithm known as Q-Learning. Our results can be the basis for more sophisticated techniques of non-contact molecular manipulations. This allows to identify and manoeuvre single molecules at will, building the basis for future bottom-up constructions of nanotechnology. [1] G. Rapenne et al., Nature Rev. Mater. 2, 17040 (2017) [2] G. J. Simpson et al., Nature Nanotech. 12, 604 (2017)