Portail HAL Ensta
Not a member yet
11080 research outputs found
Sort by
Nonlinear Propagation and Filamentation on 100 Meter Air Path of Femtosecond Beam Partitioned by Wire Mesh
No full textInternational audienceHigh-intensity (∼1 TW/cm2 and higher) region formed in the propagation of ∼60 GW, 90 fs Ti:Sapphire laser pulse on a ∼100 m path in air spans for several tens of meters and includes a plasma filament and a postfilament light channel. The intensity in this extended region is high enough to generate an infrared supercontinuum wing and to initiate laser-induced discharge in the gap between the electrodes. In the experiment and simulations, we delay the high-intensity region along the propagation direction by inserting metal-wire meshes with square cells at the laser system output. We identify the presence of a high-intensity region from the clean-spatial-mode distributions, appearance of the infrared supercontinuum wing, and occurrence of the laser-induced discharge. In the case of free propagation (without any meshes), the onset of the high-intensity zone is at 40–52 m from the laser system output with ∼30 m extension. Insertion of the mesh with 3 mm cells delays the beginning of the high-intensity region to 49–68 m with the same ∼30 m extension. A decrease in the cell size to 1 mm leads to both delay and shrinking of the high-intensity zone to 71–73 m and 6 m, respectively. Three-dimensional simulations in space confirm the mesh-induced delay of the high-intensity zone as the cell size decreases
Panaches thermiques en convection turbulente de Rayleigh-Bénard par Fluorescence induite par laser
No full textNational audienc
Social Network Structure Shapes Innovation: Experience-sharing in RL with SAPIENS
No full textThe human cultural repertoire relies on innovation: our ability to continuously and hierarchically explore how existing elements can be combined to create new ones. Innovation is not solitary, it relies on collective accumulation and merging of previous solutions. Machine learning approaches commonly assume that fully connected multi-agent networks are best suited for innovation. However, human laboratory and field studies have shown that hierarchical innovation is more robustly achieved by dynamic communication topologies. In dynamic topologies, humans oscillate between innovating individually or in small clusters, and then sharing outcomes with others. To our knowledge, the role of multi-agent topology on innovation has not been systematically studied in machine learning. It remains unclear a) which communication topologies are optimal for which innovation tasks, and b) which properties of experience sharing improve multi-level innovation. Here we use a multi-level hierarchical problem setting (WordCraft), with three different innovation tasks. We systematically design networks of DQNs sharing experiences from their replay buffers in varying topologies (fully connected, small world, dynamic, ring). Comparing the level of innovation achieved by different experience-sharing topologies across different tasks shows that, first, consistent with human findings, experience sharing within a dynamic topology achieves the highest level of innovation across tasks. Second, experience sharing is not as helpful when there is a single clear path to innovation. Third, two metrics we propose, conformity and diversity of shared experience, can explain the success of different topologies on different tasks. These contributions can advance our understanding of optimal AI-AI, human-human, and human-AI collaborative networks, inspiring future tools for fostering collective innovation in large organizations
Self-similar crack propagation along a viscoelastic interface in a double-cantilever beam test
No full textInternational audienceDouble-Cantilever Beam (DCB) testing is a common protocol to evaluate bonded interface toughness. The data-analysis procedures are initially based on the classical Linear Elastic Fracture Mechanics (LEFM) and have been extended to deal with plastic behavior. Nevertheless, those analyses are not suitable when time-dependent behavior is involved in the crack propagation process. In this paper, an analysis of crack propagation along a viscoelastic interface during a DCB test is conducted, assuming a Standard Linear Solid (SLS) model for the adhesive. During the self-similar crack propagation regime, a steady-state stress–strain distribution is achieved ahead of the crack tip and a Eulerian description is used. A finite-difference scheme is implemented to solve the set of differential equations from which stress–strain evolutions along the bondline are determined as the specimen deforms. The crack propagation response under stationary loading conditions is then simulated and the energy-based failure criteria are evaluated comparing both local and global estimations of the Strain-Energy Release Rate (SERR)
Evaluation of a smart home platform for adults with Down syndrome
No full textInternational audiencePeople with Down syndrome (DS) encounter difficulties in their daily lives. In recent years, smart homes have provided some answers to the problem of residential autonomy. In fact, smart homes can provide support for complex routines and activities, while adjusting to the person's behaviors and needs and offering maximum control of the environment. In this context, the objectives of this study were to assess (i) the usability of the smart home platform as well as (ii) the impact of the use (for 6 months) of this platform by adults with DS on many dimensions such as residential autonomy, projection into the future, the development of a life plan, self-determination, quality of life, self-esteem, anxiety, and psychological well-being. The first objective of a preliminary study was to verify that the support platform we use is suitable and can be used effectively by people with DS.For the preliminary study, the evaluations show effective, satisfactory and efficient use of all applications. Regarding the objective of our study, the evaluations show positive impacts on many dimensions such as living skills, self-determination, quality of life, self-esteem, anxiety and certain components of life psychological well-being
A high-order discontinuous Galerkin Method using a mixture of Gauss-Legendre and Gauss-Lobatto quadratures for improved efficiency
No full textIn discontinuous Galerkin spectral element methods (DGSEM), the two most common approaches to numerically integrate the terms of the weak form are either using Gauss-Legendre or Gauss-Lobatto quadratures. The former yields more accurate results but at a higher computational cost, so that a priori it is not clear whether one approach is more efficient that the other. In this paper, it is shown (theoretically for a particular case and numerically for the general case) that using Gauss-Lobatto quadrature for the convection matrix actually introduces a negligible error. In contrast, using Gauss-Lobatto quadratures for the evaluation of the jump term in the element faces introduces a sizeable error. This leads to the proposal of a new DG approach, where the convection matrix is evaluated using Gauss-Lobatto quadratures, whereas the face mass matrices are integrated using Gauss-Legendre quadratures. For elements with constant Jacobian and constant coefficients, a formal proof shows that no numerical integration error is actually introduced in the evaluation of the residual, even though both the mass and the convection matrices are not computed exactly with Gauss-Lobatto quadratures. For elements with non-constant Jacobian and/or non-constant coefficients, the impact of numerical integration error on the overall error is evaluated through a series of numerical tests, showing that this is also negligible. In addition, the computational cost associated to the matrix-vector products required to evaluate the residual is evaluated precisely for the different cases considered. The proposed approach is particularly attractive in the most general case, since the use of Gauss-Lobatto quadratures significantly speeds-up the evaluation of the residual
Egalisation hybride des signaux FTN à faible complexité pour les systèmes acoustiques sous-marins
No full textInternational audienceThe underwater acoustic channel has a limited bandwidth; it is thus necessary to optimize the spectrum efficiency with for instancethe choice of a Faster-Than-Nyquist technique. However, this necessarily leads to Inter-Symbol Interference that needs to be tackled in order toachieve a good system performance ; this becomes especially critical for the underwater acoustic channel and we thus propose an original hybridturbo equalization method for systems using Faster-Than-Nyquist signaling. Differently from the traditional Hybrid Time-Frequency DomainEqualizer (HTFDE), with the proposed method, each Frequency Domain Equalizer (FDE) is followed by a phase error estimator and an adaptiveTime Domain Equalizer (TDE). Our simulations, including real underwater channels, show that our proposal improves the system performancecompared to conventional methods
Experimental Validation of Ellipsoidal Techniques for State Estimation in Marine Applications
No full textInternational audienceA reliable quantification of the worst-case influence of model uncertainty and external disturbances is crucial for the localization of vessels in marine applications. This is especially true if uncertain GPS-based position measurements are used to update predicted vessel locations that are obtained from the evaluation of a ship's state equation. To reflect real-life working conditions, these state equations need to account for uncertainty in the system model, such as imperfect actuation and external disturbances due to effects such as wind and currents. As an application scenario, the GPS-based localization of autonomous DDboat robots is considered in this paper. Using experimental data, the efficiency of an ellipsoidal approach, which exploits a bounded-error representation of disturbances and uncertainties, is demonstrated
