Hunting active Brownian particles: Learning optimal behavior
- Publikationstyp:
- Zeitschriftenaufsatz
- Metadaten:
-
- Autoren
- Marcel Gerhard
- Ashreya Jayaram
- Andreas Fischer
- Thomas Speck
- Autoren-URL
- https://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=fis-test-1&SrcAuth=WosAPI&KeyUT=WOS:000725690400014&DestLinkType=FullRecord&DestApp=WOS_CPL
- DOI
- 10.1103/PhysRevE.104.054614
- eISSN
- 2470-0053
- Externe Identifier
- Clarivate Analytics Document Solution ID: XH8PS
- PubMed Identifier: 34942812
- ISSN
- 2470-0045
- Ausgabe der Veröffentlichung
- 5
- Zeitschrift
- PHYSICAL REVIEW E
- Artikelnummer
- ARTN 054614
- Datum der Veröffentlichung
- 2021
- Status
- Published
- Titel
- Hunting active Brownian particles: Learning optimal behavior
- Sub types
- Article
- Ausgabe der Zeitschrift
- 104
Datenquelle: Web of Science (Lite)
- Andere Metadatenquellen:
-
- Autoren
- Marcel Gerhard
- Ashreya Jayaram
- Andreas Fischer
- Thomas Speck
- DOI
- 10.1103/physreve.104.054614
- eISSN
- 2470-0053
- ISSN
- 2470-0045
- Ausgabe der Veröffentlichung
- 5
- Zeitschrift
- Physical Review E
- Sprache
- en
- Artikelnummer
- 054614
- Online publication date
- 2021
- Status
- Published online
- Herausgeber
- American Physical Society (APS)
- Herausgeber URL
- http://dx.doi.org/10.1103/physreve.104.054614
- Datum der Datenerfassung
- 2021
- Titel
- Hunting active Brownian particles: Learning optimal behavior
- Ausgabe der Zeitschrift
- 104
Datenquelle: Crossref
- Abstract
- We numerically study active Brownian particles that can respond to environmental cues through a small set of actions (switching their motility and turning left or right with respect to some direction) which are motivated by recent experiments with colloidal self-propelled Janus particles. We employ reinforcement learning to find optimal mappings between the state of particles and these actions. Specifically, we first consider a predator-prey situation in which prey particles try to avoid a predator. Using as reward the squared distance from the predator, we discuss the merits of three state-action sets and show that turning away from the predator is the most successful strategy. We then remove the predator and employ as collective reward the local concentration of signaling molecules exuded by all particles and show that aligning with the concentration gradient leads to chemotactic collapse into a single cluster. Our results illustrate a promising route to obtain local interaction rules and design collective states in active matter.
- Addresses
- Institut für Physik, Johannes Gutenberg-Universität Mainz, Staudingerweg 7-9, 55128 Mainz, Germany.
- Autoren
- Marcel Gerhard
- Ashreya Jayaram
- Andreas Fischer
- Thomas Speck
- DOI
- 10.1103/physreve.104.054614
- eISSN
- 2470-0053
- Externe Identifier
- PubMed Identifier: 34942812
- Funding acknowledgements
- Carl-Zeiss-Stiftung:
- Open access
- false
- ISSN
- 2470-0045
- Ausgabe der Veröffentlichung
- 5-1
- Zeitschrift
- Physical review. E
- Sprache
- eng
- Medium
- Paginierung
- 054614
- Datum der Veröffentlichung
- 2021
- Status
- Published
- Datum der Datenerfassung
- 2021
- Titel
- Hunting active Brownian particles: Learning optimal behavior.
- Sub types
- Journal Article
- Ausgabe der Zeitschrift
- 104
Datenquelle: Europe PubMed Central
- Abstract
- We numerically study active Brownian particles that can respond to environmental cues through a small set of actions (switching their motility and turning left or right with respect to some direction) which are motivated by recent experiments with colloidal self-propelled Janus particles. We employ reinforcement learning to find optimal mappings between the state of particles and these actions. Specifically, we first consider a predator-prey situation in which prey particles try to avoid a predator. Using as reward the squared distance from the predator, we discuss the merits of three state-action sets and show that turning away from the predator is the most successful strategy. We then remove the predator and employ as collective reward the local concentration of signaling molecules exuded by all particles and show that aligning with the concentration gradient leads to chemotactic collapse into a single cluster. Our results illustrate a promising route to obtain local interaction rules and design collective states in active matter.
- Date of acceptance
- 2021
- Autoren
- Marcel Gerhard
- Ashreya Jayaram
- Andreas Fischer
- Thomas Speck
- Autoren-URL
- https://www.ncbi.nlm.nih.gov/pubmed/34942812
- DOI
- 10.1103/PhysRevE.104.054614
- eISSN
- 2470-0053
- Ausgabe der Veröffentlichung
- 5-1
- Zeitschrift
- Phys Rev E
- Sprache
- eng
- Country
- United States
- Paginierung
- 054614
- Datum der Veröffentlichung
- 2021
- Status
- Published
- Titel
- Hunting active Brownian particles: Learning optimal behavior.
- Sub types
- Journal Article
- Ausgabe der Zeitschrift
- 104
Datenquelle: PubMed
- Beziehungen:
- Eigentum von