Dresden 2020 – wissenschaftliches Programm

Die DPG-Frühjahrstagung in Dresden musste abgesagt werden! Lesen Sie mehr ...

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

O: Fachverband Oberflächenphysik

O 25: Poster Session - Focus Sessions: Innovation in Machine learning PRocEsses for Surface Science (IMPRESS)

O 25.3: Poster

Montag, 16. März 2020, 18:15–20:00, P1A

Neural Network Controlled Nanocar — •Bernhard Ramsauer and Oliver T. Hofmann — Institute of Solid State Physics, NAWI Graz, Graz University of Technology, Petersgasse 16, 8010 Graz, Austria

In 2021 at the nanocar race at Center for Materials Development and Structure Studies (CEMES-CNRS) in Toulouse, France we are planning to participate with the world*s first neural-network-controlled nanocar. At this race, participants have to direct a single molecule across a 'race-track' set on a metallic substrate, controlling their nanocars via an STM-tip without being in physical contact with it.

Although nanocars can be readily synthesized with different shapes and properties, the physics that govern the molecule's movement is complex and involves the interaction between the molecule and the tip as well as between the molecule and the substrate. Therefore, it is far from straightforward for humans to manoeuvre the nanocar and predict the result of a performed action.

To improve the performance, we implement Deep Neural Networks (DNNs), which are able to perform various actions even for subsequently changing environmental conditions. The DNN enables an 'autonomous' driving of nanocars by controlling the STM-tip position and the applied voltage based on the nanocars position and orientation on the surface. Moreover, the DNNs will yield direct physical insight into the interaction that governs the nanocar maneuvers.