arXiv subjects

Aditya Sengupta

Publications and source records attributed to Aditya Sengupta.

3 recordsLinked to original sources

arXiv2026

Design and simulation of a photonic lantern-inspired astrophotonic chip for spectral sensing

Compact astrophotonic sensors can trade general-purpose spectral coverage for task-specific wavelength discrimination in a small integrated footprint. Developed with the Mazin Lab at UC Santa Barbara, this chip couples a single-mode input into a multimode interference region and seven-port fanout, producing wavelength-dependent output power fingerprints for spectral retrieval. Using ANSYS Lumerical FDTD simulations in silicon nitride, we compare symmetric and staggered-release geometries and quantify throughput, wavelength-dependent changes in the seven-port output distribution, and sensitivity near a 745 nm design point. We optimize over the tested geometric parameter space and identify designs with improved throughput and wavelength discrimination, including increased throughput-weighted Fisher information relative to the baseline. These results suggest that lantern-inspired integrated photonics can enable compact, task-specific spectral sensors for astrophotonic applications.

astro-ph.IM↗

arXiv2025

ArtInsight: Enabling AI-Powered Artwork Engagement for Mixed Visual-Ability Families

We introduce ArtInsight, a novel AI-powered system to facilitate deeper engagement with child-created artwork in mixed visual-ability families. ArtInsight leverages large language models (LLMs) to craft a respectful and thorough initial description of a child's artwork, and provides: creative AI-generated descriptions for a vivid overview, audio recording to capture the child's own description of their artwork, and a set of AI-generated questions to facilitate discussion between blind or low-vision (BLV) family members and their children. Alongside ArtInsight, we also contribute a new rubric to score AI-generated descriptions of child-created artwork and an assessment of state-of-the-art LLMs. We evaluated ArtInsight with five groups of BLV family members and their children, and as a case study with one BLV child therapist. Our findings highlight a preference for ArtInsight's longer, artistically-tailored descriptions over those generated by existing BLV AI tools. Participants highlighted the creative description and audio recording components as most beneficial, with the former helping ``bring a picture to life'' and the latter centering the child's narrative to generate context-aware AI responses. Our findings reveal different ways that AI can be used to support art engagement, including before, during, and after interaction with the child artist, as well as expectations that BLV adults and their sighted children have about AI-powered tools.

cs.HC↗

arXiv2023

Using the Gerchberg-Saxton algorithm to reconstruct non-modulated pyramid wavefront sensor measurements

Adaptive optics (AO) is a technique to improve the resolution of ground-based telescopes by correcting, in real-time, optical aberrations due to atmospheric turbulence and the telescope itself. With the rise of Giant Segmented Mirror Telescopes (GSMT), AO is needed more than ever to reach the full potential of these future observatories. One of the main performance drivers of an AO system is the wavefront sensing operation, consisting of measuring the shape of the above mentioned optical aberrations. Aims. The non-modulated pyramid wavefront sensor (nPWFS) is a wavefront sensor with high sensitivity, allowing the limits of AO systems to be pushed. The high sensitivity comes at the expense of its dynamic range, which makes it a highly non-linear sensor. We propose here a novel way to invert nPWFS signals by using the principle of reciprocity of light propagation and the Gerchberg-Saxton (GS) algorithm. We test the performance of this reconstructor in two steps: the technique is first implemented in simulations, where some of its basic properties are studied. Then, the GS reconstructor is tested on the Santa Cruz Extreme Adaptive optics Laboratory (SEAL) testbed located at the University of California Santa Cruz. This new way to invert the nPWFS measurements allows us to drastically increase the dynamic range of the reconstruction for the nPWFS, pushing the dynamics close to a modulated PWFS. The reconstructor is an iterative algorithm requiring heavy computational burden, which could be an issue for real-time purposes in its current implementation. However, this new reconstructor could still be helpful in the case of many wavefront control operations. This reconstruction technique has also been successfully tested on the Santa Cruz Extreme AO Laboratory (SEAL) bench where it is now used as the standard way to invert nPWFS signal.

astro-ph.IM↗