SMuK 2023 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
HK: Fachverband Physik der Hadronen und Kerne
HK 36: Computing I
HK 36.4: Vortrag
Mittwoch, 22. März 2023, 16:30–16:45, SCH/A117
A language model based tracking algorithm for the Straw Tube Tracker of the PANDA experiment — •Jakapat Kannika1,2, James Ritman1,2,3, and Tobias Stockmanns3 — 1GSI Helmholtz Centre for Heavy Ion Research, Darmstadt, Germany — 2Ruhr-Universität Bochum, Bochum, Germany — 3Forschungszentrum Jülich, Jülich, Germany
The Straw Tube Tracker (STT) is designed for momentum reconstruction of charged particles in the PANDA experiment. This talk will present a tracking algorithm that can group measured hit positions of the STT into tracks of the particles based on the language model. The overall algorithm consists of two main parts, the language model which contains the probabilities for predicting the next hit point based on previous measurements, and the tracking algorithm, the program that uses the information from the language model to select the most probable track or filter possible track candidates. We performed track parameterizations perpendicular and parallel to the solenoidal magnetic field and compared the reconstructed tracks to the MC truth information. As a result, all the reconstructed parameters are shown to be reconstructed within the expected ranges according to the MC information. The algorithm is also being developed to include a branching algorithm that can select the best track out of multiple track candidates. The development involves improving the quality of hit information and creating a track selector. The talk will also present the efficiency and resolution of this algorithm to reconstruct tracks in the STT.