The rise of large satellite constellations and Distributed Space Systems (DSS) demands generalized frameworks that enable fault-tolerant, autonomous distributed space applications. Conventional ground-centric command and control does not scale to systems of tens or hundreds of satellites, motivating the adoption of distributed computing. This paper introduces a conceptual distributed computing framework for satellite swarms, covering distributed state, command and control, and scientific mission. As a first validation step, a strongly eventually consistent distributed state service is designed and implemented using Conflict-free Replicated Data Types (CRDT), specifically...

Quantization consists of replacing the original data types used to represent the weights of neural networks with less resource-intensive data types. While considerable research has focused on quantization, most existing methods that offer theoretical guarantees do so by providing error bounds on the difference between the original and reduced precision models. In this article, we introduce a new quantization technique that, rather than focusing on bounding errors, determines the minimum precision necessary to preserve class dominance, independent of any specific set of numerical formats. In other words, regardless of the exact scores for each class, our method guarantees...

[...]

In this work, the propagation of electromagnetic waves in realistic turbulent media is studied. Combining large eddy simulation (LES) for generating the largest scales of the medium -the resolved scales -and stochastic processes for the smallest scales -the sub-grid scales -, extended LES (X-LES) allows to generate large and realistic domains of turbulent atmosphere. In this work, we focus on the size of the smallest resolved scales and its impact on an X-band electromagnetic wave propagating in the medium. This impact is studied via numerical simulations of a 10 GHz spherical wave propagating through 95 km of turbulent media generated with X-LES and for different...

Performance evaluation of mobile ad hoc network (MANET) routing protocols relies heavily on simulation, yet the choice of propagation model, a critical determinant of link connectivity, is rarely justified or systematically examined. Through a bibliometric audit of 32 comparative routing studies, we demonstrate that urban MANET evaluations systematically employ propagation models unsuited to their claimed deployment contexts, including free-space and two-ray models that ignore building obstruction entirely, and, in a majority of cases, empirical urban models such as ITU-R P.1411 that capture aggregate attenuation but not individual building geometry. We then show that...

In the context of the IPaTRo project, we aim to build a telepresence robot which allows hospitalized children to keep in touch with their families: children pilot the robot in the family homes from their beds in the hospital. Using the turtlebot4 as a base, our goal is to create a robot design that respects key guidelines: usable, accepted, replicable and low-cost. Through an iterative user-centered design process we created several design variants. We printed the resulting model with a 3D printer. With that, we currently conduct experimentations with children from age 6 to 17 to observe their reactions towards the robot and their preferences in term of customization...

<div><p>Global Navigation Satellite Systems receivers estimate their position, velocity and timing, by first measuring the signal propagation delay and Doppler shift to a set of visible satellites, and then solving a multilateration problem. However, the presence of interfering signals clearly degrades the estimation accuracy of these synchronization parameters, which requires to account for robust mitigation strategies. In this work, we further explore a recently introduced method that exploits the geometric properties of the constant modulus (CM) interference family within a modified Expectation Maximization (EM) framework, modeling CM interferences as a Riemannian...

This paper studies a new Bayesian Information Criterion (BIC) for matrix Lie groups (LGs) adapted to determine the number of components in a mixture of Gaussian distributions in LGs. A generic expression of this criterion is derived taking into account the curvature of the considered LG. Contrary to the standard Euclidean BIC, the theoretical expression of the LG BIC depends on the LG geometry. The proposed BIC is validated numerically for a clustering problem with a mixture of Gaussian distributions defined in SE(2). It successfully recovers the true number of clusters, even in scenarios where the cluster geometry is challenging to identify.

Critical embedded systems increasingly demand both autonomy and safety. Numerical optimization algorithms enable complex behaviors, but their formal verification remains an open-challenge. In this work, we present a framework building on numerical analysis of firstorder methods, by Lessard et al., to prove the convergence of such algorithms at code level. The approach expresses the algorithm as a discretetime dynamical system. Convergence is then captured as a dissipativity property, computed by solving Linear Matrix Inequalities (LMIs), and expressed as witnesses of convergence at the code level. These specifications are then provided to the weakest precondition engine...