Portail HAL Ensta
Not a member yet
11080 research outputs found
Sort by
Robustness Testing of an Industrial Road Object Detection System
No full textInternational audienceAs AI-based critical systems are expected to operate in dynamic environments, it is crucial to ensure their reliability under various operational conditions. In computer vision, one way to achieve this is by testing the system's robustness to input image perturbations. However, while many methods have been proposed and evaluated in academic settings, their effectiveness and applicability in practice remain largely unknown. In this paper, we report the results of testing the robustness of an industrial case of an AI-based road object detection system, in a black-box setting. By defining relevant perturbations and metrics, we analyse the system's response to changes in its hardware and software environment, and identify areas for improvement through retraining with data augmentation. We address the key challenges encountered during this evaluation and provide insights that may help practitioners in performing similar tests and guide future research on robustness testing of AI-based object detection systems
Experimental and modelling of the vapor-liquid equilibria of [C mim]Br(n = 2, 3, 4) + H2O systems
No full textInternational audienc
C−H Functionalization of Aldehydes and Ketones with Transient Directing Groups: Recent Developments
No full textInternational audienceIn order to directly functionalize C−H bonds of complex molecules and, in particular, to control the regioselectivity of the reaction, a wide range of directing groups has been used. However, these directing groups need to be installed and removed for further applications, which may limit the use of C−H activation in synthesis. Concerning aldehydes and ketones, a transient directing group strategy has recently emerged to overcome this drawback. The addition of an additive, in general an amine, allowed the in situ formation of the real directing group to achieve C−H activation. This review presents the latest developments in the field over the period 2020–2023
Prerequisite structure discovery for an intelligent tutoring system based on intrinsic motivation
No full textInternational audienceThis paper addresses the importance of Knowledge Structure (KS) and Knowledge Tracing (KT) in improving the recommendation of educational content in intelligent tutoring systems. The KS represents the relations between different Knowledge Components (KCs), while KT predicts a learner's success based on her past history. The contribution of this research includes proposing a KT model that incorporates the KS as a learnable parameter, enabling the discovery of the underlying KS from learner trajectories. The quality of the uncovered KS is assessed by using it to recommend content and evaluating the recommendation algorithm with simulated students
Multi-Horizon Virtual Sensor for Controllable Suspensions: A Benchmark of SOTA Deep Forecasting Models
No full textInternational audienceIn the pursuit of achieving autonomous driving, ensuring comfortable and efficient handling of the vehicle is of paramount importance. One way to meet these requirements is by utilizing controllable suspensions as an active system. However, designing an effective control strategy requires knowledge of specific dynamical states, such as suspension stroke speed and displacement. The present study proposes a multi-horizon Virtual Sensor (VS) capable of estimating these states in order to address this issue. The VS is intended to replace or enhance limited direct measurement, and it is designed to cope with predictive control systems where future finite-horizon states are necessary. Importantly, the suggested approach is model-free and based on state-of-the-art deep forecasting models. The benchmark of the chosen models was conducted based on real experimental tests and evaluated using multiple metrics. Our study improves previous work in several aspects, including the use of a minimal instrumentation setup, accurate tracking performance over multiple horizons, the ability to explain the predictions, and the provision of confidence intervals for the estimated states
A revised model of ammonium perchlorate combustion with detailed kinetics
No full textInternational audienceIn this study, we propose a revised coupled combustion model for ammonium perchlorate (AP), leveraging recent advances in the modelling of ammonia and NOx chemistry. A coupled combustion model relies on three founding bricks: a detailed gas-phase kinetic model, a condensed-phase decomposition model, and a pyrolysis law describing the relationship between the surface temperature and mass ow. The proposed gas-phase kinetic model, is validated against data on species sampling in jet-stirred reactors, laminar ame speed, and ignition delay time. These test cases, rarely used by the solid propellant community, highlight deciencies in a reference mechanism from the literature. A new model for AP decomposition in the condensed phase is proposed to be used with the gas-phase mechanism. A suitable pyrolysis law is designed using the Zel'dovich-Novozhilov theory to ensure the stability of the coupled combustion model. The methodology employed is described in detail, for others to replicate. Finally, the overall model is applied to simulate the AP laminar ame in a 1D coupled approach. These calculations provide results on the regression rate, surface temperature, temperature sensitivity and species proles for prescribed initial temperature of AP and ambient pressure. The behavior of the proposed combustion model is presented in comparison with other reference models. The role of gas-phase kinetics in modeling AP combustion is discussed
A hybrid wavelet-based method to model the electromagnetic propagation above a polluted sea surface
No full textInternational audienceElectromagnetic propagation above a rough polluted sea surface differs from one above a clean sea. Indeed, a damp effect on the waves appears. This can be used to detect a pollutant leakage. In this article, we model the propagation of electromagnetic waves above a polluted sea using a fast wavelet-based method and a two-scale model. Numerical simulations in S-band are provided
Reduced-order modeling of geometrically nonlinear rotating structures using the direct parametrisation of invariant manifolds
No full textInternational audienceThe direct parametrisation method for invariant manifolds is a nonlinear reduction technique which derives nonlinear mappings and reduced-order dynamics that describe the evolution of dynamical systems along a low-dimensional invariant-based span of the phase space. It can be directly applied to finite element problems. When the development is performed using an arbitrary order asymptotic expansion, it provides an efficient reduced-order modeling strategy for geometrically nonlinear structures. It is here applied to the case of rotating structures featuring centrifugal effect. A rotating cantilever beam with large amplitude vibrations is first selected in order to highlight the main features of the method. Numerical results show that the method provides accurate reduced-order models (ROMs) for any rotation speed and vibration amplitude of interest with a single master mode, thus offering remarkable reduction in the computational burden. The hardening/softening transition of the fundamental flexural mode with increasing rotation speed is then investigated in detail and a ROM parametrised with respect to rotation speed and forcing frequencies is detailed. The method is then applied to a twisted plate model representative of a fan blade, showing how the technique can handle more complex structures. Hardening/softening transition is also investigated as well as interpolation of ROMs, highlighting the efficacy of the method
Education numérique - Manuel Cycle 2 (5-6°) - Collection Décodage 2024
No full textOuvrage collectifInternational audienceLe manuel CODAGE d’Éducation numérique pour le Cycle 2 56 a été réalisé dans le cadre du projet cantonal d’introduction de l’Éducation numérique dans le cursus scolaire vaudois.Il est le fruit d’une collaboration entre la Direction pédagogique de la DGEO (Direction générale de l’enseignement obligatoire et de la pédagogie spécialisée) du canton de Vaud, le Centre des Sciences de l’apprentissage LEARN de l’EPFL (École polytechnique fédérale de Lausanne), la HEP Vaud (Haute école pédagogique du canton de Vaud), et l’UNIL (Université de Lausanne), avec l’expertise d’Inria (Institut français de recherche en sciences et technologies du numérique)
An Overview of the Recent Advances in Composite Materials and Artificial Intelligence for Hydrogen Storage Vessels Design
No full textInternational audienceThe environmental impact of CO2 emissions is widely acknowledged, making the development of alternative propulsion systems a priority. Hydrogen is a potential candidate to replace fossil fuels for transport applications, with three technologies considered for the onboard storage of hydrogen: storage in the form of a compressed gas, storage as a cryogenic liquid, and storage as a solid. These technologies are now competing to meet the requirements of vehicle manufacturers; each has its own unique challenges that must be understood to direct future research and development efforts. This paper reviews technological developments for Hydrogen Storage Vessel (HSV) designs, including their technical performance, manufacturing costs, safety, and environmental impact. More specifically, an up-to-date review of fiber-reinforced polymer composite HSVs was explored, including the end-of-life recycling options. A review of current numerical models for HSVs was conducted, including the use of artificial intelligence techniques to assess the performance of composite HSVs, leading to more sophisticated designs for achieving a more sustainable future