Navigation überspringen
Universitätsbibliothek Heidelberg
Standort: ---
Exemplare: ---
heiBIB
 Online-Ressource
Verfasst von:Stradmann, Yannik [VerfasserIn]   i
 Schemmel, Johannes [VerfasserIn]   i
Titel:Closing the loop
Titelzusatz:high-speed robotics with accelerated neuromorphic hardware
Verf.angabe:Yannik Stradmann and Johannes Schemmel
E-Jahr:2024
Jahr:26 March 2024
Umfang:6 S.
Illustrationen:Illustrationen
Fussnoten:Gesehen am 27.05.2024
Titel Quelle:Enthalten in: Frontiers in neuroscience
Ort Quelle:Lausanne : Frontiers Research Foundation, 2007
Jahr Quelle:2024
Band/Heft Quelle:18(2024) vom: März, Artikel-ID 1360122, Seite 1-6
ISSN Quelle:1662-453X
Abstract:<p>The BrainScaleS-2 system is an established analog neuromorphic platform with versatile applications in the diverse fields of computational neuroscience and spike-based machine learning. In this work, we extend the system with a configurable realtime event interface that enables a tight coupling of its distinct analog network core to external sensors and actuators. The 1,000-fold acceleration of the emulated nerve cells allows us to target high-speed robotic applications that require precise timing on a microsecond scale. As a showcase, we present a closed-loop setup for commuting brushless DC motors: we utilize PyTorch to train a spiking neural network emulated on the analog substrate to control an electric motor from a sensory event stream. The presented system enables research in the area of event-driven controllers for high-speed robotics, including self-supervised and biologically inspired online learning for such applications.</p>
DOI:doi:10.3389/fnins.2024.1360122
URL:kostenfrei: Volltext: https://doi.org/10.3389/fnins.2024.1360122
 kostenfrei: Volltext: https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2024.1360122/full
 DOI: https://doi.org/10.3389/fnins.2024.1360122
Datenträger:Online-Ressource
Sprache:eng
Sach-SW:analog neuromorphic computing
 Closed-loop controller
 High-speed robotics
 motor control
 neurorobotics
 spiking neural networks
K10plus-PPN:1889918121
Verknüpfungen:→ Zeitschrift
 
 
Lokale URL UB: Zum Volltext

Permanenter Link auf diesen Titel (bookmarkfähig):  https://katalog.ub.uni-heidelberg.de/titel/69217677   QR-Code
zum Seitenanfang