This title appears in the Scientific Report :
2020
Please use the identifier:
http://hdl.handle.net/2128/25706 in citations.
Please use the identifier: http://dx.doi.org/10.1126/sciadv.abb6987 in citations.
Autonomous robotic nanofabrication with reinforcement learning
Autonomous robotic nanofabrication with reinforcement learning
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-sc...
Saved in:
Personal Name(s): | Leinen, Philipp |
---|---|
Esders, Malte / Schütt, Kristof T. / Wagner, Christian (Corresponding author) / Müller, Klaus-Robert (Corresponding author) / Tautz, F. Stefan | |
Contributing Institute: |
Quantum Nanoscience; PGI-3 |
Published in: | Science advances, 6 (2020) 36, S. eabb6987 - |
Imprint: |
Washington, DC [u.a.]
Assoc.
2020
|
PubMed ID: |
32917594 |
DOI: |
10.1126/sciadv.abb6987 |
Document Type: |
Journal Article |
Research Program: |
Controlling Configuration-Based Phenomena |
Link: |
Get full text OpenAccess OpenAccess Get full text |
Publikationsportal JuSER |
Please use the identifier: http://dx.doi.org/10.1126/sciadv.abb6987 in citations.
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. |