arXiv subjects

Jack Reilly

Publications and source records attributed to Jack Reilly.

6 recordsLinked to original sources

arXiv2025

TransitReID: Transit OD Data Collection with Occlusion-Resistant Dynamic Passenger Re-Identification

Transit Origin-Destination (OD) data are fundamental for optimizing public transit services, yet current collection methods, such as manual surveys, Bluetooth/WiFi tracking, and Automated Passenger Counters, are often costly, device-dependent, or unable to support individual-level matching. Meanwhile, onboard surveillance cameras already deployed on most transit vehicles provide an underutilized opportunity for automated OD data collection. Leveraging this, we present TransitReID, a framework for individual-level and occlusion-resistant passenger re-identification (ReID) tailored to transit environments. TransitReID introduces three key components: (1) an occlusion- and viewpoint-robust ReID algorithm that integrates a variational autoencoder-guided region-attention mechanism with selective feature pooling to emphasize visible and discriminative body regions; (2) a Hierarchical Storage and Dynamic Matching (HSDM) mechanism that adapts static ReID matching to dynamic bus operations while balancing accuracy, memory, and speed; and (3) a multi-threaded edge implementation that enables near real-time OD estimation while preserving privacy through local data processing. We also construct a new Transit ReID dataset with over 17,000 images captured from real bus front/rear cameras under diverse occlusion and viewpoint conditions. Experimental results show that TransitReID achieves state-of-the-art ReID performance, attaining 88.3% R-1 accuracy on the proposed transit ReID dataset and sustaining 80-90% OD estimation accuracy in both simulations and real-world operation, with deployment supported on NVIDIA Jetson edge devices. This work provides an algorithmic and system-level foundation for scalable, privacy-preserving automated transit OD collection.

cs.CV↗

arXiv2022

A compact instrument for gamma-ray burst detection on a CubeSat platform II: Detailed design, assembly and validation

The Gamma-ray Module, GMOD, is a miniaturised novel gamma-ray detector which will be the primary scientific payload on the Educational Irish Research Satellite (EIRSAT-1) 2U CubeSat mission. GMOD comprises a compact (25mm $\times$ 25mm $\times$ 40mm) cerium bromide scintillator coupled to a tiled array of 4$\times$4 silicon photomultipliers, with front-end readout provided by the IDE3380 SIPHRA. This paper presents the detailed GMOD design and the accommodation of the instrument within the restrictive CubeSat form factor. The electronic and mechanical interfaces are compatible with many off-the-shelf CubeSat systems and structures. The energy response of the GMOD engineering qualification model has been determined using radioactive sources, and an energy resolution of 5.4% at 662keV has been measured. EIRSAT-1 will perform on-board processing of GMOD data. Trigger results, including light-curves and spectra, will be incorporated into the spacecraft beacon and transmitted continuously. Inexpensive hardware can be used to decode the beacon signal, making the data accessible to a wide community. GMOD will have scientific capability for the detection of gamma-ray bursts, in addition to the educational and technology demonstration goals of the EIRSAT-1 mission. The detailed design and measurements to date demonstrate the capability of GMOD in low Earth orbit, the scalability of the design for larger CubeSats and as an element of future large gamma-ray missions.

astro-ph.IM↗

arXiv2021

Embedded Firmware Development for a Novel CubeSat Gamma-Ray Detector

The Gamma-ray Module (GMOD) is an experiment designed for the detection of gamma-ray bursts in low Earth orbit as the principal scientific payload on a 2-U CubeSat, EIRSAT-1. GMOD comprises a cerium bromide scintillator coupled to silicon photomultipliers which are processed and digitised by a bespoke ASIC. Custom firmware on the GMOD motherboard has been designed, implemented and tested for the MSP430 microprocessor which manages the experiment including readout, storage and configuration of the system. The firmware has been verified in a series of experiments testing the response over a realistic range of input detector trigger frequencies from 50Hz to 1kHz for the primary time tagged event (TTE) data. The power consumption and ability of the firmware to successfully receive and transmit the packets to the on-board computer was investigated. The experiment demonstrated less than 1% loss of packets up to 1kHz for the standard transfer mode with the power not exceeding 31mW. The transfer performance and power consumption demonstrated are within the required range of this CubeSat instrument.

astro-ph.IM↗

arXiv2020

Assembly, Integration, and Verification Activities for a 2U CubeSat, EIRSAT-1

The Educational Irish Research Satellite, EIRSAT-1, is a project developed by students at University College Dublin that aims to design, build, and launch Ireland's first satellite. EIRSAT-1 is a 2U CubeSat incorporating three novel payloads; GMOD, a gamma-ray detector, EMOD, a thermal coating management experiment, and WBC, a novel attitude control algorithm. The EIRSAT-1 project is carried out with the support of the Education Office of the European Space Agency, under the educational Fly your Satellite! programme. The Assembly, Integration and Verification plan for EIRSAT-1 is central to the philosophy and the development of the spacecraft. The model philosophy employed for the project is known as the 'prototype' approach in which two models of the spacecraft are assembled; an Engineering Qualification Model (EQM) and a Flight Model (FM). The payloads, GMOD and EMOD, and the Antenna Deployment Module (ADM) platform element warrant a Development Model in addition to an EQM and a FM, as they have been designed and developed in-house. After successful completion of the Critical Design Review and Ambient Test Readiness Review phases of the project, the EQM of EIRSAT-1 will be assembled and integrated. After assembly and integration of the EQM, the project will begin the ambient test campaign, in which the EQM undergoes ambient functional and mission testing. This work details the preparation and execution of the assembly, integration, and verification activities of EIRSAT-1 EQM.

astro-ph.IM↗

arXiv2020

Flight Software Development for the EIRSAT-1 Mission

The Educational Irish Research Satellite, known as EIRSAT-1, is a student-led project to design, build, test and launch Ireland's first satellite. The on-board software for this mission is being developed using Bright Ascension's GenerationOne Flight Software Development Kit. This paper provides an overview of this kit and of EIRSAT-1's on-board software design. Drawing on the team's contrasting experience with writing entirely custom firmware for the mission's science payloads, this work discusses the impact of using a kit on the software development process. The challenges associated with the educational nature of this project are the focus of this discussion. The objective of this paper is to provide useful information for other CubeSat teams assessing software development options.

astro-ph.IM↗

arXiv2013

Arriving on time: estimating travel time distributions on large-scale road networks

Most optimal routing problems focus on minimizing travel time or distance traveled. Oftentimes, a more useful objective is to maximize the probability of on-time arrival, which requires statistical distributions of travel times, rather than just mean values. We propose a method to estimate travel time distributions on large-scale road networks, using probe vehicle data collected from GPS. We present a framework that works with large input of data, and scales linearly with the size of the network. Leveraging the planar topology of the graph, the method computes efficiently the time correlations between neighboring streets. First, raw probe vehicle traces are compressed into pairs of travel times and number of stops for each traversed road segment using a `stop-and-go' algorithm developed for this work. The compressed data is then used as input for training a path travel time model, which couples a Markov model along with a Gaussian Markov random field. Finally, scalable inference algorithms are developed for obtaining path travel time distributions from the composite MM-GMRF model. We illustrate the accuracy and scalability of our model on a 505,000 road link network spanning the San Francisco Bay Area.

cs.LG↗