Göttingen 2025 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

T: Fachverband Teilchenphysik

T 97: Data, AI, Computing, Electronics IX (AI-based Object Reconstruction)

T 97.4: Vortrag

Freitag, 4. April 2025, 09:45–10:00, VG 2.102

GCNN-based Hybrid Reconstruction of Cosmic Rays with IceAct and IceCube — •Lars Marten, Philipp Behrens, Shuyang Deng, Lasse Düser, Jonas Häußler, Lars Heuermann, Sönke Schwirn, Philipp Soldin, Julian Vogt, and Christopher Wiebusch — RWTH Aachen - III. physikalisches Institut B, Aachen, Germany

IceAct is an array of Imaging Air Cherenkov Telescopes stationed at the South Pole as part of the IceCube Neutrino Observatory. Among its goals is the combined measurement of air showers together with the in-ice detector IceCube and the IceTop surface detector. Such hybrid measurements grant the advantage of complementary information improving reconstruction capabilities. Our graph convolutional neural network has been developed using a simulation of the IceAct array with the purpose of reconstructing the direction of the primary particle of an air shower. In this talk we will present an updated version of this neural network with additional reconstruction capabilities such as primary particle energy and relative shower core position. Also, we will present our advances in including hybrid data into our network prediction.

Keywords: Imaging Air Cherenkov Telescope; IceAct; cosmic-ray induced air-showers; hybrid measurements; graph convolutional neural network