001041610 001__ 1041610
001041610 005__ 20250424202216.0
001041610 0247_ $$2doi$$a10.48550/ARXIV.2002.11952
001041610 037__ $$aFZJ-2025-02344
001041610 1001_ $$0P:(DE-Juel1)164154$$aLeinen, Philipp$$b0
001041610 245__ $$aAutonomous robotic nanofabrication with reinforcement learning
001041610 260__ $$barXiv$$c2020
001041610 3367_ $$0PUB:(DE-HGF)25$$2PUB:(DE-HGF)$$aPreprint$$bpreprint$$mpreprint$$s1745493949_28179
001041610 3367_ $$2ORCID$$aWORKING_PAPER
001041610 3367_ $$028$$2EndNote$$aElectronic Article
001041610 3367_ $$2DRIVER$$apreprint
001041610 3367_ $$2BibTeX$$aARTICLE
001041610 3367_ $$2DataCite$$aOutput Types/Working Paper
001041610 520__ $$aThe ability to handle single molecules as effectively as macroscopic building-blocks would enable the construction of complex supramolecular structures inaccessible to self-assembly. The fundamental challenges obstructing this goal are the uncontrolled variability and poor observability of atomic-scale conformations. Here, we present a strategy to work around both obstacles, and demonstrate autonomous robotic nanofabrication by manipulating single molecules. Our approach employs reinforcement learning (RL), which finds solution strategies even in the face of large uncertainty and sparse feedback. We demonstrate the potential of our RL approach by removing molecules autonomously with a scanning probe microscope from a supramolecular structure -- an exemplary task of subtractive manufacturing at the nanoscale. Our RL agent reaches an excellent performance, enabling us to automate a task which previously had to be performed by a human. We anticipate that our work opens the way towards autonomous agents for the robotic construction of functional supramolecular structures with speed, precision and perseverance beyond our current capabilities.
001041610 536__ $$0G:(DE-HGF)POF4-5213$$a5213 - Quantum Nanoscience (POF4-521)$$cPOF4-521$$fPOF IV$$x0
001041610 588__ $$aDataset connected to DataCite
001041610 650_7 $$2Other$$aMesoscale and Nanoscale Physics (cond-mat.mes-hall)
001041610 650_7 $$2Other$$aArtificial Intelligence (cs.AI)
001041610 650_7 $$2Other$$aMachine Learning (cs.LG)
001041610 650_7 $$2Other$$aRobotics (cs.RO)
001041610 650_7 $$2Other$$aFOS: Physical sciences
001041610 650_7 $$2Other$$aFOS: Computer and information sciences
001041610 7001_ $$0P:(DE-HGF)0$$aEsders, Malte$$b1
001041610 7001_ $$0P:(DE-HGF)0$$aSchütt, Kristof T.$$b2
001041610 7001_ $$0P:(DE-Juel1)140276$$aWagner, Christian$$b3$$eCorresponding author$$ufzj
001041610 7001_ $$0P:(DE-Juel1)5055$$aMüller, Klaus-Robert$$b4$$eCorresponding author
001041610 7001_ $$0P:(DE-Juel1)128791$$aTautz, F. Stefan$$b5$$ufzj
001041610 773__ $$a10.48550/ARXIV.2002.11952
001041610 8564_ $$uhttps://arxiv.org/abs/2002.11952
001041610 909CO $$ooai:juser.fz-juelich.de:1041610$$pVDB
001041610 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)140276$$aForschungszentrum Jülich$$b3$$kFZJ
001041610 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)128791$$aForschungszentrum Jülich$$b5$$kFZJ
001041610 9131_ $$0G:(DE-HGF)POF4-521$$1G:(DE-HGF)POF4-520$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5213$$aDE-HGF$$bKey Technologies$$lNatural, Artificial and Cognitive Information Processing$$vQuantum Materials$$x0
001041610 9201_ $$0I:(DE-Juel1)PGI-3-20110106$$kPGI-3$$lQuantum Nanoscience$$x0
001041610 980__ $$apreprint
001041610 980__ $$aVDB
001041610 980__ $$aI:(DE-Juel1)PGI-3-20110106
001041610 980__ $$aUNRESTRICTED