Portail HAL Ensta
Not a member yet
11080 research outputs found
Sort by
Task and motion planning methods: applications and limitations
No full textInternational audienceRobots are required to perform more and more complicated tasks, which raises the requirement of more intelligent planning algorithms. As a domain having been explored for decades, task and motion planning (TAMP) methods have achieved significant results, but several challenges remain to be solved. This paper summarizes the development of TAMP from solving objectives, simulation environments, methods and remaining limitations. In particular, it compares different simulation environments and methods used in different tasks aiming to provide a practical guide and overview for the beginners
L’intelligence artificielle et le développement des technologies vertes : le cas de la propulsion électrique et des systèmes de recharge
No full textInternational audiencePour de nombreux chercheurs souhaitant étudier l’impact de l’intelligence artificielle (IA) sur les processus d'innovation technologique, il est essentiel de disposer d’une mesure robuste de sa diffusion. Même s’il ne persiste que peu de doute sur le potentiel ubiquitaire élevé de l’IA, elle n’en est pas moins aujourd'hui qu'à son enfance, et la mesure ex ante de la nouveauté radicale reste l’un des plus grands défis à relever pour les analystes du changement technique. Nous proposons, dans cette contribution, une telle mesure utilisant des données de brevets, mesure sur la base de laquelle nous élaborons des cartographies technologiques qui cherchent à rendre compte de la manière dont s’imbriquent effectivement IA et technologies vertes, et dont cette imbrication se transforme dans le temps. Nous sommes alors à même de repérer des trajectoires technologiques et leurs inflexions, voire d'éventuelles ruptures paradigmatiques. Nous comparons ces dynamiques pour trois technosystèmes spécifiques : le véhicule hybride, le véhicule électrique et les systèmes de recharge de ces véhicules. Nous constatons empiriquement que les domaines technologiques de l’électrification des véhicules concernés par l’IA se sont sensiblement diversifiés ces dernières années
Effet de l'inertie sur le flambement d'une poutre sous un chargement axial dynamique
No full textInternational audienc
Réseau d’antennes In-Band Full-Duplex Ultra-Large-Bande à double polarisation
No full textNational audienceLe réseau d’antennes Ultra-Large-Bande présenté ici est constitué de 4 antennes à double polarisation linéaire alimentées spécifiquement pour créer une structuration symétrique nécessaire à l’obtention d’un fort niveau d’isolation (SIC) en champ proche entre les parties émission et réception. Ce système constitue ainsi l’étage antennaire d’un dispositif « In-Band Full-Duplex » à même d’émettre et de recevoir simultanément sur la même bande de fréquences. Deux autres innovations ont également été développées pour cette application : 2 baluns de 1 vers 4 pour alimenter les parties RX et TX et une nouvelle méthode d’alimentation identique et symétrique pour les deux polarisations. Les simulations du système complet montrent une bonne adaptation de 2 à 20 GHz avec un gain effectif moyen de 11 dBi et une annulation de l’auto-interférence de 120dB en simulation et attendue à 40-50dB en pratique
Mechanical Properties of Ultra-high Molecular Weight Polyethylene (tensylon) from Tensile Tests
No full textInternational audienceResults on mechanical properties of Tensylon composites at room temperature are presented. Single-ply and two-ply samples, obtained from the ply-precursor sheet (of two orthogonal layers) have been subjected to: load till failure in traction, at different strain rates (below 10-1 s-1) and cycles of successive loading and unloading and 5 min stress relaxation period. The characteristic times of relaxation are evaluated and the difference in values of Young modulus before and after the relaxation stage is established. A complex cyclic/relaxation test requires a visco-elasto-plastic model of Tensylon, and allows to quantify it. This model predicts the material behavior in other types of tests: for instance, it predicts strain rate independence of loading to failure in the considered strain rate range. Cyclic tests fulfilled at a fixed strain rate suggest that Tensylon is an elastoplastic material without noticeable viscosity. The proposed model, additively including nonlinear viscoelasticity and plastic flow with strengthening, shows a satisfactory agreement with experimental data. It also agrees that the material is strain-rate-insensitive in the range 10-3 s-1-10-1 s-1
Experimental and numerical investigation of drop weight impact of aramid and UHMWPE reinforced epoxy
No full textInternational audienceAbstract Due to their characteristics such as weight/strength ratio and absorbed energy, the widespread use of composite materials in the last decades engorged the companies to exploit these materials invariant applications like the aerospace, automobile, and marine hull. However, there are some obstructs to the use of these materials that may constrain that. This came from the fact, that composite materials suffer from different damages modes that occur during loading and can be lead to catastrophic failure in their structure, such as intralaminar and interlaminar damage. Consequently, this motivated the researchers to study its behavior considering different damage modes and at different loading states. This work performed a finite element simulation using the Abaqus program of low-velocity drop impact for epoxy reinforced with Kevlar 49 and Ultra High Molecular Weight Polyethylene (UHMWPE) with different thicknesses and number of layers. A user-defined material VUMAT subroutine-based progressive damage model, and the Hashin failure criteria implemented in Abaqus Explicit finite element code had been utilized in this work. In Addition, the interlaminar damage models depend on the cohesive zone model (CZM). The numerical simulation results were compared with the experiments data to confirm the reliability of the numerical model
Distributed Algorithms for Scalable Proximity Operator Computation and Application to Video Denoising
No full textInternational audienceOptimization problems arising in signal and image processing involve an increasingly large number of variables. In addition to the curse of dimensionality, another difficulty to overcome is that the cost function usually reads as the sum of several loss/regularization terms, which are non-necessarily smooth and possibly composed with large-size linear operators. Proximal splitting approaches are fundamental tools to address such problems, with demonstrated efficiency in many applicative fields. In this paper, we present a new distributed algorithm for computing the proximity operator of a sum of non-necessarily smooth convex functions composed with arbitrary linear operators. Our algorithm relies on a primal-dual splitting strategy, and benefits from established convergence guaranties. Each involved function is associated with a node of a hypergraph, with the ability to communicate with neighboring nodes sharing the same hyperedge. Thanks to this structure, our method can be efficiently implemented on modern parallel computing architectures, distributing the computations on multiple nodes or machines, with controlled requirements for synchronization steps. Good numerical performance and scalability properties are demonstrated on a problem of video sequence denoising. Our code implemented in Julia is made available at \url{https://github.com/MarinENSTA/distributed_julia_denoising}
Outliers and uncertainties in GNSS ZTD estimates from double-difference processing and precise point positioning
No full textInternational audienc
Waveform Control of Relativistic Electron Dynamics in Laser-Plasma Acceleration
No full textInternational audienceThe interaction of ultraintense laser pulses with an underdense plasma is used in laser-plasma acceleration to create compact sources of ultrashort pulses of relativistic electrons and x rays. The accelerating structure is a plasma wave, or wakefield, that is excited by the laser ponderomotive force, a force that is usually assumed to depend solely on the laser envelope and not on its exact waveform. Here, we use near-single-cycle laser pulses with a controlled carrier-envelope phase to show that the actual waveform of the laser field has a clear impact on the plasma response. The beam pointing of our relativistic electron beam oscillates in phase with the carrier-envelope phase of the laser, at an amplitude of 15 mrad, or 30% of the beam divergence. Numerical simulations explain this observation through asymmetries in the injection and acceleration of the electron beam, which are locked to the carrier-envelope phase. These results imply that we achieve waveform control of relativistic electron dynamics. Our results pave the way to high-precision, subcycle control of electron injection in plasma accelerators, enabling the production of attosecond relativistic electron bunches and x rays