TY  - JOUR
AU  - Leinen, Philipp
AU  - Esders, Malte
AU  - Schütt, Kristof T.
AU  - Wagner, Christian
AU  - Müller, Klaus-Robert
AU  - Tautz, F. Stefan
TI  - Autonomous robotic nanofabrication with reinforcement learning
JO  - Science advances
VL  - 6
IS  - 36
SN  - 2375-2548
CY  - Washington, DC [u.a.]
PB  - Assoc.
M1  - FZJ-2020-03214
SP  - eabb6987 -
PY  - 2020
AB  - The ability to handle single molecules as effectively as macroscopic building blocks would enable the construction of complex supramolecular structures inaccessible to self-assembly. The fundamental challenges obstructing this goal are the uncontrolled variability and poor observability of atomic-scale conformations. Here, we present a strategy to work around both obstacles and demonstrate autonomous robotic nanofabrication by manipulating single molecules. Our approach uses reinforcement learning (RL), which finds solution strategies even in the face of large uncertainty and sparse feedback. We demonstrate the potential of our RL approach by removing molecules autonomously with a scanning probe microscope from a supramolecular structure. Our RL agent reaches an excellent performance, enabling us to automate a task that previously had to be performed by a human. We anticipate that our work opens the way toward autonomous agents for the robotic construction of functional supramolecular structures with speed, precision, and perseverance beyond our current capabilities.
LB  - PUB:(DE-HGF)16
C6  - pmid:32917594
UR  - <Go to ISI:>//WOS:000567766700027
DO  - DOI:10.1126/sciadv.abb6987
UR  - https://juser.fz-juelich.de/record/884720
ER  -