Szabolcs Velkei | Searcharxiv

arXiv2025

NEO and imminent impactor discoveries from Hungary: recent results and lessons learnt

2022 EB5, 2023 CX1 and 2024 BX1: these are the three recent imminent impactor discoveries from the Piszkéstető Mountain Station of the Konkoly Observatory. They make up about one percent of all NEO discoveries from our observatory and here we provide a detailed description of our approach and methodology that led to this noticeable observational sensitivity to these meter-sized impactors. After outlining the historical background of astronomical discoveries from Hungary, we introduce our recently upgraded survey instrumentation and outline the observational strategy and its implementation. We highlight the importance of strong feedback between analysis and ongoing data collection, maximizing the value of immediate follow-up. Finally, we discuss plans for moving forward to increase the sensitivity and the temporal coverage of our survey.

astro-ph.EP↗

arXiv2025

A large-scale, physically-based synthetic dataset for satellite pose estimation

The Deep Learning Visual Space Simulation System (DLVS3) introduces a novel synthetic dataset generator and a simulation pipeline specifically designed for training and testing satellite pose estimation solutions. This work introduces the DLVS3-HST-V1 dataset, which focuses on the Hubble Space Telescope (HST) as a complex, articulated target. The dataset is generated using advanced real-time and offline rendering technologies, integrating high-fidelity 3D models, dynamic lighting (including secondary sources like Earth reflection), and physically accurate material properties. The pipeline supports the creation of large-scale, richly annotated image sets with ground-truth 6-DoF pose and keypoint data, semantic segmentation, depth, and normal maps. This enables the training and benchmarking of deep learning-based pose estimation solutions under realistic, diverse, and challenging visual conditions. The paper details the dataset generation process, the simulation architecture, and the integration with deep learning frameworks, and positions DLVS3 as a significant step toward closing the domain gap for autonomous spacecraft operations in proximity and servicing missions.

cs.CV↗