arXiv subjects

Robert Kennedy

Publications and source records attributed to Robert Kennedy.

6 recordsLinked to original sources

arXiv2024

Prithvi-EO-2.0: A Versatile Multi-Temporal Foundation Model for Earth Observation Applications

This paper presents Prithvi-EO-2.0, a new geospatial foundation model that offers significant improvements over its predecessor, Prithvi-EO-1.0. Trained on 4.2 million global time series samples from NASA's Harmonized Landsat and Sentinel-2 data archive at 30-m resolution, the new model incorporates temporal and location embeddings for enhanced performance across various geospatial tasks. Through extensive benchmarking with GEO-Bench, the model outperforms the previous Prithvi-EO model by 8% across a range of tasks. It also outperforms six other geospatial foundation models when benchmarked on remote sensing tasks from different domains and resolutions (i.e. from 0.1 m to 15 m). The results demonstrate the versatility of the model in both classical Earth observation and high-resolution applications. Early involvement of end-users and subject matter experts (SMEs) allowed constant feedback on model and dataset design, enabling customization across diverse SME-led applications in disaster response, land cover and crop mapping, and ecosystem dynamics monitoring. Prithvi-EO-2.0 is available as an open-source model on Hugging Face and IBM TerraTorch, with additional resources on GitHub. The project exemplifies the Trusted Open Science approach embraced by all involved organizations.

cs.CV↗

arXiv2021

Interstellar Now! Missions to and Sample Returns from Nearby Interstellar Objects

The recently discovered first high velocity hyperbolic objects passing through the Solar System, 1I/'Oumuamua and 2I/Borisov, have raised the question about near term missions to Interstellar Objects. In situ spacecraft exploration of these objects will allow the direct determination of both their structure and their chemical and isotopic composition, enabling an entirely new way of studying small bodies from outside our solar system. In this paper, we map various Interstellar Object classes to mission types, demonstrating that missions to a range of Interstellar Object classes are feasible, using existing or near-term technology. We describe flyby, rendezvous and sample return missions to interstellar objects, showing various ways to explore these bodies characterizing their surface, dynamics, structure and composition. Interstellar objects likely formed very far from the solar system in both time and space; their direct exploration will constrain their formation and history, situating them within the dynamical and chemical evolution of the Galaxy. These mission types also provide the opportunity to explore solar system bodies and perform measurements in the far outer solar system.

astro-ph.IM↗

arXiv2021

Interstellar Objects in the Solar System: 1. Isotropic Kinematics from the Gaia Early Data Release 3

1I/'Oumuamua (or 1I) and 2I/Borisov (or 2I), the first InterStellar Objects (ISOs) discovered passing through the solar system, have opened up entirely new areas of exobody research. Finding additional ISOs and planning missions to intercept or rendezvous with these bodies will greatly benefit from knowledge of their likely orbits and arrival rates. Here, we use the local velocity distribution of stars from the Gaia Early Data Release 3 Catalogue of Nearby Stars and a standard gravitational focusing model to predict the velocity dependent flux of ISOs entering the solar system. With an 1I-type ISO number density of $\sim$0.1 AU$^{-3}$, we predict that a total of $\sim$6.9 such objects per year should pass within 1 AU of the Sun. There will be a fairly large high-velocity tail to this flux, with half of the incoming ISOs predicted to have a velocity at infinity, v$_{\infty}$, $>$ 40 km s$^{-1}$. Our model predicts that $\sim$92\% of incoming ISOs will be residents of the galactic thin disk, $\sim$6\% ($\sim$4 per decade) will be from the thick disk, $\sim$1 per decade will be from the halo and at most $\sim$3 per century will be unbound objects, ejected from our galaxy or entering the Milky Way from another galaxy. The rate of ISOs with very low v$_{\infty}$ $\lesssim$ 1.5 km s$^{-1}$ is so low in our model that any incoming very low velocity ISOs are likely to be previously lost solar system objects. Finally, we estimate a cometary ISO number density of $\sim$7 $\times$ 10$^{-5}$ AU$^{-3}$ for 2I type ISOs, leading to discovery rates for these objects possibly approaching once per decade with future telescopic surveys.

astro-ph.EP↗

arXiv2020

Fluorescence-Guided Raman Spectroscopy for Tumour Margin Delineation

The initial detection and identification of suspicious lesions and the precise delineation of tumour mar-gins are essential for successful tumour resection, with progression-free survival linked to rates of complete resection. However, post-surgical positive margin rates remain high for many cancers and despite numerous advances in intraoperative imaging and diagnostic technologies, there exists no single modality that can adequately perform both tumoural detection and delineation. Here, we demonstrate a multimodal computer vision-based diagnostic system capable of both the gross detection and identification of suspicious lesions and the precise delineation of disease margins. We first show that through visual tracking of a spectroscopic probe, we enable real-time tumour margin delineation both for ex vivo human tumour biopsies and for an in vivo tumour xenograft mouse model. We then demonstrate that the combination of Raman spectroscopic diagnoses with protoporphyrin IX (PPIX) fluorescence imaging enables fluorescence-guided Raman spectroscopic margin delineation. Our fluorescence-guided Raman spectroscopic system achieves superior margin delineation accuracy to fluorescence imaging alone, demonstrating the potential for our system to achieve improved clinical outcomes for tumour resection surgeries.

physics.med-ph↗

arXiv2020

Exobodies in Our Back Yard: Science from Missions to Nearby Interstellar Objects

The recent discovery of the first confirmed Interstellar Objects (ISOs) passing through the Solar System on clearly hyperbolic objects opens the potential for near term ISO missions, either to the two known objects, or to similar objects found in the future. Such ISOs are the only exobodies we have a chance of accessing directly in the near future. This White Paper focuses on the science possible from in situ spacecraft exploration of nearby ISOs. Such spacecraft missions are technically possible now and are suitable potential missions in the period covered by the 2023-2032 Decadal Survey. Spacecraft missions can determine the structure and the chemical and isotopic composition of ISO in a close flyby coupled with a small sub-probe impactor and either a mass spectrometer or a high resolution UV spectrometer; this technology will also be useful for fast missions to TransNeptune Objects (TNOs) and long period comets. ISO exploration holds the potential of providing considerable improvements in our knowledge of galactic evolution, of planetary formation, and of the cycling of astrobiologically important materials through the galaxy.

astro-ph.IM↗

arXiv2017

HEPCloud, a New Paradigm for HEP Facilities: CMS Amazon Web Services Investigation

Historically, high energy physics computing has been performed on large purpose-built computing systems. These began as single-site compute facilities, but have evolved into the distributed computing grids used today. Recently, there has been an exponential increase in the capacity and capability of commercial clouds. Cloud resources are highly virtualized and intended to be able to be flexibly deployed for a variety of computing tasks. There is a growing nterest among the cloud providers to demonstrate the capability to perform large-scale scientific computing. In this paper, we discuss results from the CMS experiment using the Fermilab HEPCloud facility, which utilized both local Fermilab resources and virtual machines in the Amazon Web Services Elastic Compute Cloud. We discuss the planning, technical challenges, and lessons learned involved in performing physics workflows on a large-scale set of virtualized resources. In addition, we will discuss the economics and operational efficiencies when executing workflows both in the cloud and on dedicated resources.

cs.DC↗