arXiv subjects

Nicholas Collins

Publications and source records attributed to Nicholas Collins.

4 recordsLinked to original sources

arXiv2026

Nancy Grace Roman Space Telescope Wide Field Instrument: Bright Point Source Saturation Response and Persistence Properties from Thermal-Vacuum Testing

The Nancy Grace Roman Space Telescope's Wide Field Instrument (WFI) will observe hundreds of thousands of bright stars across its Core Community Surveys, particularly in the dense stellar fields of the Galactic Bulge Time Domain Survey (GBTDS). Sources brighter than ~17th magnitude will saturate WFI detector pixels in typical survey exposures, with the brightest stars deeply saturating large pixel regions and potentially producing persistence signals that may impact subsequent observations. Prior detector characterization did not explore the regime of deep point source saturation. To address this gap, we conducted a bright star saturation test during WFI's second Thermal Vacuum test campaign (TVAC2) at BAE Space & Mission Systems in Boulder, CO. Using the Stimulus of Ray Cones (SORC) telescope simulator, we projected nine in-focus point sources through the F146 filter onto two Sensor Chip Assemblies (SCAs), with fluxes tuned to approximate stellar magnitudes ranging from ~4 to ~18 in ~170 s exposures. We present analyses of the saturation response and persistence properties of these detectors. We find that the saturated region of a ~4 mag source grows to ~150 pixels in diameter after ~170 s of illumination, compared to ~15 pixels for a ~12 mag source. Pixels adjacent to the expanding saturation front exhibit pronounced non-linear behavior consistent with charge leakage from saturated neighbors. For persistence, we find that the median signal in the first post-illumination dark exposure is broadly consistent across source magnitudes spanning ~4 to ~17, and that persistence decays to detector background levels ($\lesssim$0.05 e$^{-}$ s$^{-1}$) within approximately 20 minutes, consistent with flat field persistence measurements from the same TVAC2 campaign. These pre-flight characterization results inform community understanding of WFI detector response to prepare for Roman science.

astro-ph.IM↗

arXiv2022

Near-Surface Electrical Characterisation of Silicon Electronic Devices Using Focused keV Ions

The demonstration of universal quantum logic operations near the fault-tolerance threshold establishes ion-implanted near-surface donor atoms as a plausible platform for scalable quantum computing in silicon. The next technological step forward requires a deterministic fabrication method to create large-scale arrays of donors, featuring few hundred nanometre inter-donor spacing. Here, we explore the feasibility of this approach by implanting low-energy ions into silicon devices featuring an enlarged 60x60 $\mu$m sensitive area and an ultra-thin 3.2 nm gate oxide - capable of hosting large-scale donor arrays. By combining a focused ion beam system incorporating an electron-beam-ion-source with in-vacuum ultra-low noise ion detection electronics, we first demonstrate a versatile method to spatially map the device response characteristics to shallowly implanted 12 keV $^1$H$_2^+$ ions. Despite the weak internal electric field, near-unity charge collection efficiency is obtained from the entire sensitive area. This can be explained by the critical role that the high-quality thermal gate oxide plays in the ion detection response, allowing an initial rapid diffusion of ion induced charge away from the implant site. Next, we adapt our approach to perform deterministic implantation of a few thousand 24 keV $^{40}$Ar$^{2+}$ ions into a predefined micro-volume, without any additional collimation. Despite the reduced ionisation from the heavier ion species, a fluence-independent detection confidence of $\geq$99.99% was obtained. Our system thus represents not only a new method for mapping the near-surface electrical landscape of electronic devices, but also an attractive framework towards mask-free prototyping of large-scale donor arrays in silicon.

cond-mat.mes-hall↗

arXiv2019

Partially Observable Planning and Learning for Systems with Non-Uniform Dynamics

We propose a neural network architecture, called TransNet, that combines planning and model learning for solving Partially Observable Markov Decision Processes (POMDPs) with non-uniform system dynamics. The past decade has seen a substantial advancement in solving POMDP problems. However, constructing a suitable POMDP model remains difficult. Recently, neural network architectures have been proposed to alleviate the difficulty in acquiring such models. Although the results are promising, existing architectures restrict the type of system dynamics that can be learned --that is, system dynamics must be the same in all parts of the state space. TransNet relaxes such a restriction. Key to this relaxation is a novel neural network module that classifies the state space into classes and then learns the system dynamics of the different classes. TransNet uses this module together with the overall architecture of QMDP-Net[1] to allow solving POMDPs that have more expressive dynamic models, while maintaining efficient data requirement. Its evaluation on typical benchmarks in robot navigation with initially unknown system and environment models indicates that TransNet substantially out-performs the quality of the generated policies and learning efficiency of the state-of-the-art method QMDP-Net.

cs.RO↗

arXiv2007

The Stability of the Point Spread Function of the Advanced Camera for Surveys on the Hubble Space Telescope and Implications for Weak Gravitational Lensing

(abridged) We examine the spatial and temporal stability of the HST ACS Wide Field Camera (WFC) point spread function (PSF) using the two square degree COSMOS survey. We show that stochastic aliasing of the PSF necessarily occurs during `drizzling'. This aliasing is maximal if the output pixel scale is equal to the input pixel scale of 0.05''. We show that this source of PSF variation can be significantly reduced by choosing a Gaussian drizzle kernel and by setting the output pixel size to 0.03''. We show that the PSF is temporally unstable, most likely due to thermal fluctuations in the telescope's focus. We find that the primary manifestation of this thermal drift in COSMOS images is an overall slow periodic focus change. Using a modified version of TinyTim, we create undistorted stars in a 30x30 grid across the ACS WFC CCDs. These PSF models are created for telescope focus values in the range -10 microns to +5 microns, thus spanning the allowed range of telescope focus values. We then use the approximately ten well measured stars in each COSMOS field to pick the best-fit focus value for each field. The TinyTim model stars are then used to perform PSF corrections for weak lensing allowing systematics due to incorrectly modeled PSFs to be greatly reduced. We have made the software for PSF modeling using our modified version of TinyTim available to the astronomical community. We show the effects of Charge Transfer Efficiency (CTE) degradation, which distorts the object in the readout direction, mimicking a weak lensing signal. We derive a parametric correction for the effect of CTE on the shapes of objects in the COSMOS field as a function of observation date, position within the ACS WFC field, and object flux.

astro-ph↗