arXiv subjects

Stefano Silvestrini

Publications and source records attributed to Stefano Silvestrini.

3 recordsLinked to original sources

arXiv2026

On the Geometry of Learned Representations in Event-Based Multi-Modal Egomotion Estimation

Classical approaches to event-based egomotion estimation, including those adopted by the top-performing teams of the ELOPE challenge, rely on geometric optimization frameworks such as contrast maximization, homography estimation, or dense optical flow combined with analytic motion inversion. This work investigates the geometric structure that emerges inside a multi-modal network for egomotion estimation. Event tensors, inertial measurements, and range signals are fused through a cross-modal attention architecture and trained in a batch setting. We analyze the latent space geometry and attention dynamics, showing that (i) embeddings lie on low-dimensional manifolds aligned with motion variables, (ii) attention weights adapt with angular excitation and visual reliability, and (iii) the fused representation recovers classical observability cues. These results bridge analytical estimation theory and modern data-driven fusion.

cs.CV↗

arXiv2025

Energy efficiency analysis of Spiking Neural Networks for space applications

While the exponential growth of the space sector and new operative concepts ask for higher spacecraft autonomy, the development of AI-assisted space systems was so far hindered by the low availability of power and energy typical of space applications. In this context, Spiking Neural Networks (SNN) are highly attractive due to their theoretically superior energy efficiency due to their inherently sparse activity induced by neurons communicating by means of binary spikes. Nevertheless, the ability of SNN to reach such efficiency on real world tasks is still to be demonstrated in practice. To evaluate the feasibility of utilizing SNN onboard spacecraft, this work presents a numerical analysis and comparison of different SNN techniques applied to scene classification for the EuroSAT dataset. Such tasks are of primary importance for space applications and constitute a valuable test case given the abundance of competitive methods available to establish a benchmark. Particular emphasis is placed on models based on temporal coding, where crucial information is encoded in the timing of neuron spikes. These models promise even greater efficiency of resulting networks, as they maximize the sparsity properties inherent in SNN. A reliable metric capable of comparing different architectures in a hardware-agnostic way is developed to establish a clear theoretical dependence between architecture parameters and the energy consumption that can be expected onboard the spacecraft. The potential of this novel method and his flexibility to describe specific hardware platforms is demonstrated by its application to predicting the energy consumption of a BrainChip Akida AKD1000 neuromorphic processor.

cs.NE↗

arXiv2020

Sky visibility analysis for astrophysical data return maximization in HERMES constellation

HERMES is a scientific mission composed of 3U nano-satellites dedicated to the detection and localization of high-energy astrophysical transients, with a distributed space architecture to form a constellation in Earth orbits. The space segment hosts novel miniaturized detectors to probe the X-ray temporal emission of bright events, such as Gamma-Ray Bursts (GRBs), and the electromagnetic counterparts of Gravitational Wave Events (GWEs), playing a crucial role in future multi-messenger astrophysics. During operations, at least three instruments, separated by a minimum distance shall observe a common area of the sky to perform a triangulation of the observed event. An effective detection by the nano-satellite payload is achieved by guaranteeing a beneficial orbital and pointing configuration of the constellation. The design has to cope with the limitations imposed by small space systems, such as the lack of on-board propulsion and the reduced systems budgets. The paper describes the methodologies and the proposed strategies to overcome the mission limitations, while achieving a satisfactory constellation visibility of the sky throughout the mission duration. The mission design makes use of a high-fidelity orbit propagator, combined with an innovative mission analysis tool that estimates the scientific performances of the constellation. The influence of the natural relative motion, which is crucial to achieve an effective constellation configuration without on-board orbit control, is assessed. The presented methodology can be easily extended to any kind of distributed scientific space applications, as well as to constellations dedicated to Earth and planetary observation. In addition, the visibility tool is applicable in the context of the constellation flight dynamics operations, yielding optimized results and pointing plans based on actual satellite orbital positions.

astro-ph.IM↗