Artificial Neural Networks-based Track Fitting of Cosmic Muons through Stacked Resistive Plate Chambers

TL;DR

This paper demonstrates that artificial neural networks improve the accuracy of cosmic muon track reconstruction in resistive plate chambers compared to traditional straight line fitting, especially in noisy conditions.

Contribution

The paper introduces a novel application of ANN for track fitting in RPC detectors, showing enhanced efficiency over conventional methods.

Findings

ANN outperforms straight line fit in track reconstruction efficiency

Noise impacts detection and reconstruction performance

Simulation framework effectively evaluates ANN performance

Abstract

The India-based Neutrino Observatory (INO) collaboration, as part of its detector R\&D program, has developed prototype stacks of resistive plate chambers (RPCs) to study their performance. These stacks have also been used as testbenches for the development of related hardware and software. A crucial parameter in the characterisation of these detectors and other physics studies is the detection efficiency, which is estimated from track fitting of cosmic muons passing through the stack. So far, a simple straight line fit was used for track fitting, which was sensitive to noise hits and led to rejection of events. In this paper, we present our first results of using artificial neural networks (ANN) for track fitting of cosmic muons traversing a stack of RPCs. We present in detail, the simulation framework designed for this purpose and show that ANN offers better track reconstruction efficiency than straight line fitting. We also discuss the influence of noise and detection efficiency of cosmic muons on the track reconstruction efficiency.