Position reconstruction using deep learning for the HERD PSD beam test

TL;DR

This paper demonstrates that deep learning significantly improves position resolution in the HERD PSD detector compared to traditional methods, enhancing cosmic-ray and gamma-ray measurements in space-based experiments.

Contribution

It introduces a deep learning approach for position reconstruction in the HERD PSD, achieving superior resolution over classic methods.

Findings

Deep learning method achieves 2 mm position resolution.

Deep learning outperforms classic dual-readout ratio method.

Enhanced position accuracy benefits cosmic-ray and gamma-ray detection.

Abstract

The High Energy cosmic-Radiation Detection (HERD) facility is a dedicated high energy astronomy and particle physics experiment planned to be installed on the Chinese space station, aiming to detect high-energy cosmic rays (GeV $\sim$ PeV) and high-energy gamma rays ($>$ 500 MeV). The Plastic Scintillator Detector (PSD) is one of the sub-detectors of HERD, with its main function of providing real-time anti-conincidence signals for gamma-ray detection and the secondary function of measuring the charge of cosmic-rays. In 2023, a prototype of PSD was developed and tested at CERN PS&SPS. In this paper, we investigate the position response of the PSD using two reconstruction algorithms: the classic dual-readout ratio and the deep learning method (KAN & MLP neural network). With the latter, we achieved a position resolution of 2 mm (1$\sigma$), which is significantly better than the classic method.