Archive ouverte HAL-LAAS
Not a member yet
12189 research outputs found
Sort by
Runtime Monitoring for AI-based Systems: An application to runway detection
No full textInternational audienceAutomated runway detection is a rapidly growing domain in avionics. Recent advancements in the machine learning (ML) community have enabled the development of cutting-edge neural networks (NN) for this specific task. However, given the critical nature of automated flight operations, the reliability and safety of these systems must be rigorously ensured. Runtime monitoring approaches have emerged as promising solutions to enhance these properties. Monitors are components designed to continuously supervise the behaviour and outputs of NNs during runtime. They are generally categorised into two main types: rule-based and data-driven. Rule-based monitors verify that safety properties are satisfied, making them suitable for scenarios with well-defined safety requirements. In contrast, data-driven monitors are trained on antecedents to learn patterns of erroneous or unexpected behaviour. This poster explores the main challenges and preliminary solutions for designing and evaluating such monitors. Main challenges would include (1) Which types of threats should the monitor address? (2) Which kind of error detection mechanism should be used? and (3) How should the monitor be evaluated with respect to overall system safety
Towards high-quality GaInAsN(Sb) alloys for space multi-junction solar cells grown by Molecular Beam Epitaxy
No full textInternational audienceGaInAsN alloys are particularly interesting for lattice-matched multi-junction solar cells (MJSCs) for spaceapplications. Owing to the alloy nature of this compound, one can select independently the bandgap of GaInx AsNyand its lattice constant. In high-efficiency multi-junction solar cells for space applications, it allows to increasethe number of junctions from 3 to 4 (GaInP2 - GaAs - GaInAsN - Ge) while choosing the added subcell bandgapto perfectly match Shockley-Quiesser’s efficiency optimum and keeping the structure lattice-matched to GaAs,reaching up to 41% efficienc
Fabrication of a Knudsen micropump for integration in a Pulsed tube Cryocooler
No full textInternational audienceCryogenic temperatures below 150 K are essential for advanced applications such as superconducting systems and infrared sensors. Pulsed tube cryocoolers (PTCC) offer a compact, cryogen-free solution for achieving these temperatures, but their reliance on pressure oscillations poses challenges for miniaturization. To address this, we propose the integration of a Knudsen pump as a vibration-free compressor for a miniaturized PTCC. The Knudsen pump operates using thermal transpiration within nanochannels, eliminating the need for moving parts. However, highpressure operation for PTCCs requires sub-micrometer channels and dense parallel arrays, posing fabrication and thermal management challenges. In this work, high-aspect-ratio nanochannels are fabricated in borosilicate glass using Bessel beam laser processing, achieving around 300,000 channels on a 1 cm 2 area. Metal heaters and silicon-based heat sinks are utilized to enhance the thermal gradient. Two prototype samples are fabricated and characterized via microscopy and SEM, with flow performance characterization planned
Dopage par épitaxie localisée de nanostructures pour les dispositifs de puissance en GaN
No full textInternational audienceLa fabrication de structures nanométriques innovantes est un enjeu majeur pour le développement de composants électroniques avancés en GaN. Cet article présente un procédé de réalisation de nano-caissons en P-GaN, reposant sur la technologie etch-and-regrowth. Combinant gravure et croissance épitaxiale localisée, cette méthode constitue une alternative prometteuse à l’implantation ionique pour la fabrication de diodes PN enGaN. Ce travail explore les paramètres critiques de fabrication et analyse les défis topologiques des structures. Les résultats révèlent que la géométrie et l’orientation des motifs ont un impact déterminant sur la qualité des structures. L’optimisation des processus de gravure chimique et la maîtrise des impuretés apparaissent comme des leviers essentiels pour améliorer les performances
Finite Convergence of the Moment-SOS Hierarchy on the Product of Spheres
No full textInternational audienceWe study the polynomial optimization problem of minimizing a multihomogeneous polynomial over the product of spheres. This polynomial optimization problem models the tensor optimization problem of finding the best rank one approximation of an arbitrary tensor. We show that the moment-SOS hierarchy has finite convergence in this case, for a generic multihomogeneous objective function. To show finite convergence of the hierarchy, we use a result of Huang et al. [SIAM J. Optim. 34(4) (2024), pp 3399-3428], which relies on local optimality conditions. To prove that the local optimality conditions hold generically, we use techniques from differential geometry and Morse theory. This work generalizes the main result of Huang [Optim. Lett. 17(5) (2023), pp 1263-1270], which shows finite convergence for the case of a homogeneous polynomial over a single sphere
Global optimization of low-rank polynomials
No full textThis work considers polynomial optimization problems where the objective admits a lowrank canonical polyadic tensor decomposition. We introduce LRPOP (low-rank polynomial optimization), a new hierarchy of semidefinite programming relaxations for which the size of the semidefinite blocks is determined by the canonical polyadic rank rather than the number of variables. As a result, LRPOP can solve low-rank polynomial optimization problems that are far beyond the reach of existing sparse hierarchies. In particular, we solve problems with up to thousands of variables with total degree in the thousands. Numerical conditioning for problems of this size is improved by using the Bernstein basis. The LRPOP hierarchy converges from below to the global minimum of the polynomial under standard assumptions
Backscattering and Wireless Power Transfer in Battery-Free Sensors for Structural Health Monitoring of Reinforced Concrete Structures
No full textInternational audienceThis paper presents an overview of the use of backscattering and Wireless Power Transfer in battery-free sensors for Non-Destructive Testing of reinforced concrete in order to ensure the Structural Health Monitoring of the structure of which it is the main material. A focus will be on a generic, battery-free Sensing Node. This last meets the Simultaneous Wireless Information and Power Transfer paradigm by being wirelessly powered and communicating wirelessly -simultaneously-thanks to electromagnetic waves. The security of wireless communications is also enhanced by backscattering and encrypting the power supply wave. This Sensing Node has been used in a Cyber-Physical System, and deployed and tested in a reinforced concrete beam with a power supply range of several metres and a wireless communication range -in Bluetooth Low Energy or LoRaWAN-of at least a few tens of metres
Data Informativity for Analysis and Design of Positive Systems
No full textInternational audienceThis paper studies data informativity of positive systems using linear programming (LP). The concept called data informativity represents the sufficiency of a given dataset to solve analysis/design problems. We provide the necessary and sufficient conditions for the data-driven analysis and design problems of positive systems to be solvable. Moreover, we clarify that these conditions are characterized by LP problems. We provide numerical examples to demonstrate the effectiveness of our approaches
Variation of Dynamic ON-Resistance in SiC and p-GaN HEMTs During DHTOL Operation
No full textInternational audienceWide-bandgap devices offer higher efficiency, faster switching, and better thermal performance than traditional silicon transistors. Their ability to operate at high voltages with lower losses enables compact and efficient power electronics. Since these devices are expected to find application in high/speed and high/performance power converters, their reliability is subject of intense investigation. Particular focus is on the stability of the main parameters, such as ON-resistance (Rds, on), during switching operation. To assess Rds, on stability, this paper discusses the development of a dynamic high-temperature operating life (DHTOL) setup, designed to mimic actual application conditions. The setup utilizes a boost converter to achieve higher stress voltage with a lower input power supply. It consists of two detachable boards: a daughterboard, which contains only the device under test (DUT), and a main board, which contains all the essential components of the converter and clamping circuit. The detachable design facilitates further degradation analysis of the DUTs without disordering the DUT. A thermocouple is mounted close to the thermal pad of the DUT to monitor temperature, which can be used for thermally activated degradation analysis; this is particularly useful, since we did not employ a external heat sink, but just an active cooling (fan). Furthermore, the setup is used to analyze the stability of two different 650 V commercial technologies with similar parameters: trench-based Silicon Carbide (SiC) and p-Gallium Nitride (p-GaN) High Electron Mobility Transistors (HEMTs). The results indicate that the Rds, on of the trench-based SiC device remains stable, whereas the p-GaN HEMT device exhibits a substantial increase in Rds, on after approximately 8 hours at ~50°C
Exploring Extracellular Vesicle–Associated Plasma Cell-Free DNA Structure for Agnostic Biomarker Identification
No full textNational audienceCell-free DNA (cfDNA) is a promising biomarker, offering both disease-specific insights (e.g., mutations) and systemic, agnostic information (e.g., fragmentomics)1. Yet, despite its expanding clinical use, the physicochemical properties and in vivo kinetics of cfDNA remain poorly defined within the nuclease-rich plasma environment. A central and debated hypothesis is that extracellular vesicles (EVs) transport and protect cfDNA, but the evidence has been conflicting, dominated by conditioned-media experiments, and constrained by the inherently low abundance of cfDNA in plasma2,3. Resolving the nature of cfDNA carriers is therefore critical to understanding the biology of circulating nucleic acids and unlocking their full diagnostic potential.In this study, we present an analytical workflow for plasma fractionation based on size, density, or combined size–density parameters to dissect the molecular interactions of cfDNA. Applying this approach to healthy individuals, we find that cfDNA is not associated with EVs, but instead is found within small, non-vesicular particles whose density scales with cfDNA fragment size. This observation challenges the prevailing EV-mediated transport model and provides a new lens through which to examine cfDNA stability and turnover in vivo, for example through dynamic interactions with plasma proteins and lipoproteins.Extending this workflow to samples from cancer patients, we uncover evidence of altered cfDNA structural organization, suggesting that disease states reshape the molecular packaging of circulating DNA. These findings not only redefine the physical milieu of cfDNA in plasma but also highlight new avenues for biomarker discovery, with the potential to refine early detection and patient monitoring strategies.(1)Thierry, Cancer Metastasis Rev, 2016(2)Jeppesen, Cell, 2019(3)Fernando, PLOS ONE, 201