Detailed knowledge of the radiation environment in space is an indispensable prerequisite for space missions in low Earth orbit and beyond. The RadMap Telescope is a compact radiation monitor that can characterize the radiation environment aboard spacecraft and determine the biologically relevant dose received by astronauts. Its main sensor is a tracking calorimeter made from 1024 scintillating-plastic fibers of alternating orientation and silicon photomultipliers. It allows the three-dimensional tracking and identification of cosmic-ray nuclei by measurement of their energy-deposition profiles.
The properties of nuclei traversing the detector are reconstructed using a neural-network-based analysis framework. In this contribution, we describe the three consecutive convolutional networks that we use to determine the track parameters, charge, and initial kinetic energy of each nucleus as well as the challenges of a network-based analysis approach. We demonstrate the capabilities of our framework with networks trained and evaluated on simulated data and show that the achieved performance is in agreement with the requirements of radiation monitoring. Finally, we discuss the significance of our results and the limitations of both the analysis framework and the detector.