Search

Your search keyword '"Lefebvre, Martin"' showing total 191 results
Start Over Author "Lefebvre, Martin"
191 results on '"Lefebvre, Martin"'

51. Sur une approche multi-échelle de l’influence de la vitesse sur le comportement mécanique d’un composite biosourcé Lin/Elium

Author

Lefebvre, Martin, Laboratoire des sciences de l'ingénieur, de l'informatique et de l'imagerie (ICube), École Nationale du Génie de l'Eau et de l'Environnement de Strasbourg (ENGEES)-Université de Strasbourg (UNISTRA)-Institut National des Sciences Appliquées - Strasbourg (INSA Strasbourg), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Les Hôpitaux Universitaires de Strasbourg (HUS)-Centre National de la Recherche Scientifique (CNRS)-Matériaux et Nanosciences Grand-Est (MNGE), Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Institut National de la Santé et de la Recherche Médicale (INSERM)-Institut de Chimie du CNRS (INC)-Centre National de la Recherche Scientifique (CNRS)-Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Institut National de la Santé et de la Recherche Médicale (INSERM)-Institut de Chimie du CNRS (INC)-Centre National de la Recherche Scientifique (CNRS)-Réseau nanophotonique et optique, Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Centre National de la Recherche Scientifique (CNRS)-Université de Strasbourg (UNISTRA)-Centre National de la Recherche Scientifique (CNRS), Université de Strasbourg, and Nadia Bahlouli

Subjects
Strain rate sensitivity, Fibre de lin, Flax fiber, Multi-échelle, Biobased composite, [CHIM.MATE]Chemical Sciences/Material chemistry, [SPI.MECA]Engineering Sciences [physics]/Mechanics [physics.med-ph], Composite biosourcé, Sensibilité vitesse de déformation, Multi-scale
Abstract

Numerous barriers remain to develop biobased composites in mechanical parts with high requirements (aeronautics, automotive, etc.). The prime objective of this thesis concerns the multi-scale study of the composite and its flax fiber reinforcement and to study its strain rate sensitivity at various scales. The specific application targeted is a long range drone. These researches are motivated by a lack of bibliographical data on the strain rate sensitivity of this material and non-exhaustive existing behavior models. The flax fiber alone already has a composite microstructure with a complex behavior, different from a synthetic fiber and which deserves a particular characterization. This work allows to understand and characterize the influence of this microstructure on the behavior of the composite at the macroscopic scale and to propose adequate homogenization models. Moreover, a constitutive model taking into account the strain rate sensitivity at the composite scale is proposed over a broad range of speeds from quasi-static to dynamic.; De nombreux verrous persistent pour développer des composites biosourcés dans des pièces mécaniques à hauts niveaux d’exigences (aéronautique, automobile, etc.). L’objectif principal de cette thèse concerne ainsi l’étude multi-échelle du composite et de son renfort en fibres de lin et d’étudier sa sensibilité à la vitesse de déformation à différentes échelles. L’application spécifique visée est un drone longue endurance. Ces recherches sont motivées par un manque de données bibliographiques sur la sensibilité à la vitesse de ce matériau et des modèles de comportement existants non-exhaustifs. La fibre de lin seule a déjà une microstructure composite avec un comportement complexe, différent d’une fibre synthétique et qui mérite ainsi une caractérisation particulière. Ces travaux permettent ainsi de comprendre et de caractériser l’influence de cette microstructure sur le comportement du composite à l’échelle macroscopique et de proposer des modèles d’homogénéisation adéquats. De plus, un modèle constitutif prenant en compte la sensibilité à la vitesse de déformation à l’échelle du composite est proposé sur une large plage de vitesses du quasi-statique au dynamique.

Published
2022

52. Frequency based strategy for hybrid-powered unmannedaerial vehicle

Author

Pavot, Thomas, Lefebvre, Martin, Kiefer, Renaud, Mesbahi, Tedjani, Laroche, Edouard, Laboratoire des sciences de l'ingénieur, de l'informatique et de l'imagerie (ICube), École Nationale du Génie de l'Eau et de l'Environnement de Strasbourg (ENGEES)-Université de Strasbourg (UNISTRA)-Institut National des Sciences Appliquées - Strasbourg (INSA Strasbourg), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Les Hôpitaux Universitaires de Strasbourg (HUS)-Centre National de la Recherche Scientifique (CNRS)-Matériaux et Nanosciences Grand-Est (MNGE), Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Institut National de la Santé et de la Recherche Médicale (INSERM)-Institut de Chimie du CNRS (INC)-Centre National de la Recherche Scientifique (CNRS)-Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Institut National de la Santé et de la Recherche Médicale (INSERM)-Institut de Chimie du CNRS (INC)-Centre National de la Recherche Scientifique (CNRS)-Réseau nanophotonique et optique, Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Centre National de la Recherche Scientifique (CNRS)-Université de Strasbourg (UNISTRA)-Centre National de la Recherche Scientifique (CNRS), Institut d'Electronique du Solide et des Systèmes (InESS), Université Louis Pasteur - Strasbourg I-Centre National de la Recherche Scientifique (CNRS), InESS, and Centre National de la Recherche Scientifique (CNRS)

Subjects
[SPI.NANO]Engineering Sciences [physics]/Micro and nanotechnologies/Microelectronics
Published
2022

59. A 0.9-nA Temperature-Independent 565-ppm/°C Self-Biased Current Reference in 22-nm FDSOI

Author

UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique, Lefebvre, Martin, Flandre, Denis, Bol, David, UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique, Lefebvre, Martin, Flandre, Denis, and Bol, David

Abstract

Temperature-independent current references operating in the nA range are rarely area-efficient due to the use of large resistors occupying a significant silicon area at this current level. In this paper, we introduce a nA-range constant-with-temperature (CWT) current reference relying on a self-cascode MOSFET (SCM), biased by a proportional-to-absolute-temperature voltage with a CWT offset voltage. The proposed reference has been fabricated in a 22-nm fully-depleted silicon-on-insulator (FDSOI) technology and, as a result of using an SCM, occupies a silicon area of 0.0132 mm² at least 4× smaller than state-of-the-art CWT references operating in the same current range. It consumes 5.8 nW at 0.9 V and achieves a 0.9-nA current with a line sensitivity of 0.39 %/V and a temperature coefficient of 565 ppm/°C.

Published
2022

60. A 1-to-4b 16.8-POPS/W 473-TOPS/mm2 6T-based In-Memory Computing SRAM in 22nm FD-SOI with Multi-Bit Analog Batch-Normalization

Author

UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique, Kneip, Adrian, Lefebvre, Martin, Verecken, Julien, Bol, David, UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique, Kneip, Adrian, Lefebvre, Martin, Verecken, Julien, and Bol, David

Abstract

Computing in-memory (CIM) is rapidly becoming an enticing solution to accelerate convolutional neural networks (CNNs) at the edge. Yet, low-precision current-based CIM-SRAMs face severe SNR degradation due to numerous analog non-idealities and high quantization noise when performing analog-to-digital conversion prior to digital batch-normalization (DBN). In this paper, we propose a dual-supply 1-to-4b CIM-SRAM macro in 22nm FD-SOI using 6T foundry bitcells, co-designed with a CIM-aware CNN training framework to overcome these challenges. The macro includes a multi-bit analog BN (ABN) unit combined with self-calibrating dual-phase sense-amplifiers (SCDP-SAs). Measurement results show peak 1b-normalized power and area efficiencies of 16.8POPS/W and 473TOPS/mm 2 at O.4/0.8V supply and 100 MHz, surpassing existing low-precision designs.

Published
2022

62. Contributors

Author

Bagusat, Frank, Bahlouli, Nadia, Bauer, Steffen, Baxevanakis, Konstantinos P., Cadoni, E., Chapman, David, Chen, Weinong, Couque, Hervé, Dattelbaum, Dana M., De Focatiis, Davide S.A., Demarty, Yaël, Eakins, Daniel, Fensin, Saryu J., Forquin, P., Gálvez, Francisco, Gao, Jinling, Gray, George T., III, Guo, Yazhou, Hiermaier, Stefan, Hokka, Mikko, Hou, Bing, Isakov, Matti, Jones, David R., Kuokkala, Veli-Tapani, Lauro, Franck, Lefèbvre, Martin, Li, Jianguo, Li, Xing, Li, Yulong, Mohr, Dirk, Nieto-Fuentes, Juan Carlos, Notta-Cuvier, Delphine, Peroni, Lorenzo, Rittel, Daniel, Roth, Christian C., Sancho, Rafael, Sauer, Martin, Scapin, Martina, Silberschmidt, Vadim V., Siviour, Clive Richard, Song, Bo, Stergiou, Theodosios, Suo, Tao, Verleysen, Patricia, Walley, Stephen M., Wu, Bangbiao, Xia, Kaiwen, Xu, Ying, and Yao, Wei

Published
2024
Full Text
View/download PDF

65. A 0.2-to-3.6TOPS/W Programmable Convolutional Imager SoC with In-Sensor Current-Domain Ternary-Weighted MAC Operations for Feature Extraction and Region-of-Interest Detection

Author

Lefebvre, Martin, Moreau, Ludovic, Dekimpe, Rémi, Bol, David, and UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique

Subjects
Convolutional codes, Wearable computers, Feature extraction, Solid state circuits, Analog memory, Task analysis, Smart phones
Abstract

Mixed-signal vision chips are becoming increasingly popular for low-power embedded computer vision applications on smartphones, wearables and IoT nodes, as they meet stringent power and area constraints while maintaining a sufficient level of accuracy for low- to medium-level image processing tasks. On the one hand, in-sensor processing [1, 2] enables massively parallel operation but relies on pixel-level processing elements that degrade the pixel pitch and restrict the convolutional receptive field to neighboring pixels [1], precluding multi-scale operation. On the other hand, near-sensor processing [3–5] can operate at multiple scales by pixel down sampling [3] or binning [4] but entails significant power and area overhead as an analog memory is required to store pixel values awaiting processing. In addition, previous near-sensor processing SoCs are generally application-specific and thus suffer from limited versatility. In this paper, we present a 65nm QQVGA convolutional imager SoC code named SleepSpotter capable of versatile feature extraction and region-of-interest (RoI) detection based on in-sensor current-domain MAC operations. It operates at 6 different scales, features programmable filter size (F), stride (S), and ternary filter weights (1.5b). It reaches a minimum energy of 2.5pJ/pixel•frame•filter and a peak efficiency of 3.6TOPS/W, with 29% pixel area overhead for enabling the convolution and without the need for an analog memory.

Published
2021

66. A 0.2-to-3.6TOPS/W Programmable Convolutional Imager SoC with In-Sensor Current-Domain Ternary-Weighted MAC Operations for Feature Extraction and Region-of-Interest Detection

Author

UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique, Lefebvre, Martin, Moreau, Ludovic, Dekimpe, Rémi, Bol, David, UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique, Lefebvre, Martin, Moreau, Ludovic, Dekimpe, Rémi, and Bol, David

Abstract

Mixed-signal vision chips are becoming increasingly popular for low-power embedded computer vision applications on smartphones, wearables and IoT nodes, as they meet stringent power and area constraints while maintaining a sufficient level of accuracy for low- to medium-level image processing tasks. On the one hand, in-sensor processing [1, 2] enables massively parallel operation but relies on pixel-level processing elements that degrade the pixel pitch and restrict the convolutional receptive field to neighboring pixels [1], precluding multi-scale operation. On the other hand, near-sensor processing [3–5] can operate at multiple scales by pixel down sampling [3] or binning [4] but entails significant power and area overhead as an analog memory is required to store pixel values awaiting processing. In addition, previous near-sensor processing SoCs are generally application-specific and thus suffer from limited versatility. In this paper, we present a 65nm QQVGA convolutional imager SoC code named SleepSpotter capable of versatile feature extraction and region-of-interest (RoI) detection based on in-sensor current-domain MAC operations. It operates at 6 different scales, features programmable filter size (F), stride (S), and ternary filter weights (1.5b). It reaches a minimum energy of 2.5pJ/pixel•frame•filter and a peak efficiency of 3.6TOPS/W, with 29% pixel area overhead for enabling the convolution and without the need for an analog memory.

Published
2021

67. Learning Without Feedback: Fixed Random Learning Signals Allow for Feedforward Training of Deep Neural Networks

Author

UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique, Frenkel, Charlotte, Lefebvre, Martin, Bol, David, UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique, Frenkel, Charlotte, Lefebvre, Martin, and Bol, David

Abstract

While the backpropagation of error algorithm enables deep neural network training, it implies (i) bidirectional synaptic weight transport and (ii) update locking until the forward and backward passes are completed. Not only do these constraints preclude biological plausibility, but they also hinder the development of low-cost adaptive smart sensors at the edge, as they severely constrain memory accesses and entail buffering overhead. In this work, we show that the one-hot-encoded labels provided in supervised classification problems, denoted as targets, can be viewed as a proxy for the error sign. Therefore, their fixed random projections enable a layerwise feedforward training of the hidden layers, thus solving the weight transport and update locking problems while relaxing the computational and memory requirements. Based on these observations, we propose the direct random target projection (DRTP) algorithm and demonstrate that it provides a tradeoff between accuracy and computational cost that is suitable for adaptive edge computing devices.

Published
2021

68. SleepRider: a 5.5µW/MHz Cortex-M4 MCU in 28nm FD-SOI with ULP SRAM, Biomedical AFE and Fully-Integrated Power, Clock and Back-Bias Management

Author

UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique, Dekimpe, Rémi, Schramme, Maxime, Lefebvre, Martin, Kneip, Adrian, Saeidi, Roghayeh, Xhonneux, Mathieu, Moreau, Ludovic, Gonzalez Gonzalez, Marco Antonio, Pirson, Thibault, Bol, David, UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique, Dekimpe, Rémi, Schramme, Maxime, Lefebvre, Martin, Kneip, Adrian, Saeidi, Roghayeh, Xhonneux, Mathieu, Moreau, Ludovic, Gonzalez Gonzalez, Marco Antonio, Pirson, Thibault, and Bol, David

Abstract

Ultra-low-power microcontrollers (ULP MCUs) face a performance trade-off between energy-efficient computing during activity periods and low sleep power, associated with limited wake-up time and energy. Adaptive back-biasing in FD-SOI, along with near-threshold operation at ultra-low voltage, has brought significant improvements by dynamically shifting the minimum energy point (MEP) along the frequency axis. This work introduces a highly-integrated 64-MHz ULP Cortex-M4 MCU with 96-kB SRAM in 28nm FD-SOI. A clock and power management unit (CPMU) generates all internal supplies and clocks from a 1.8-V supply, while unified frequency and back-bias regulation (UFBBR) performs PVT compensation. Custom 16-kB ULP SRAMs achieve low read/write access energy, 1.2/0.84pJ/32-bit access respectively, and provide 0.98nW/kB ultra-low-leakage data retention. A low-power biomedical analog front-end enables biopotential monitoring. The MEP is 5.5µW/MHz (8.2µW/MHz including conversion losses). Sleep power is 7.7µW with retention of logic state and 32-kB memory.

Published
2021

69. Learning Without Feedback: Fixed Random Learning Signals Allow for Feedforward Training of Deep Neural Networks

Author

Frenkel, Charlotte, Lefebvre, Martin, Bol, David, Frenkel, Charlotte, Lefebvre, Martin, and Bol, David

Abstract

While the backpropagation of error algorithm enables deep neural network training, it implies (i) bidirectional synaptic weight transport and (ii) update locking until the forward and backward passes are completed. Not only do these constraints preclude biological plausibility, but they also hinder the development of low-cost adaptive smart sensors at the edge, as they severely constrain memory accesses and entail buffering overhead. In this work, we show that the one-hot-encoded labels provided in supervised classification problems, denoted as targets, can be viewed as a proxy for the error sign. Therefore, their fixed random projections enable a layerwise feedforward training of the hidden layers, thus solving the weight transport and update locking problems while relaxing the computational and memory requirements. Based on these observations, we propose the direct random target projection (DRTP) algorithm and demonstrate that it provides a tradeoff between accuracy and computational cost that is suitable for adaptive edge computing devices.

Published
2021

70. IMPACT: A 1-to-4b 813-TOPS/W 22-nm FD-SOI Compute-in-Memory CNN Accelerator Featuring a 4.2-POPS/W 146-TOPS/mm2 CIM-SRAM With Multi-Bit Analog Batch-Normalization

Author

Kneip, Adrian, Lefebvre, Martin, Verecken, Julien, and Bol, David

Abstract

Amid a strife for ever-growing AI processing capabilities at the edge, compute-in-memory (CIM) SRAMs involving current-based dot-product (DP) operators have become excellent candidates to execute low-precision convolutional neural networks (CNNs) with tremendous energy efficiency. Yet, these architectures suffer from noticeable analog non-idealities and a lack of dynamic range adaptivity, leading to significant information loss during ADC quantization that hinders CNN performance with digital batch-normalization (DBN). To overcome these issues, we present IMPACT, a 1-to-4b mixed-signal accelerator in 22-nm FD-SOI intended for low-precision edge CNNs. It includes a novel 72-kB dual-supply CIM-SRAM macro with 6T-based DP operators as well as a multi-bit analog batch-normalization (ABN) unit to bypass the ADC quantization issue. IMPACT embeds the macro within a highly parallel, channel- and precision-adaptive digital datapath that handles memory transfers and provides input-reshaping capability. Finally, a co-designed CIM-aware CNN training framework accounts for the macro’s analog impairments, wherein non-linearity and variability. Measurement results showcase a 4b-normalized computing efficiency of 813 TOPS/W at 64 MHz for the whole accelerator. Taken aside, the CIM-SRAM macro achieves a peak energy efficiency and area efficiency of 4.2 POPS/W and 146 TOPS/mm2, respectively, surpassing all existing low-precision CIM designs to date.

Published
2023
Full Text
View/download PDF

71. Examination of Institutional Investment in the United States of America from 1999 to 2018

Author

Lefebvre, Martin R

Subjects
Behavioral Economics, Gravity Model of Trade, LDA, Institutional Investment, New York, Human Geography, Spatial Science, United States, Space-Time Analysis, Other Business, Regional Economics, Financial Geography, Business Intelligence, Latent Dirichlet Allocation, Other Geography, Business Administration, Management, and Operations, Finance
Abstract

This thesis examines the evolution of spatial preference of institutional investors located in the United States of America for the time period of 1999 to 2018 using a mix of exploratory data analysis techniques and more sophisticated space-time and machine learning techniques such as ESRI Space-Time cube and Latent Dirichlet Allocation topic modeling. This thesis concludes that despite having the appearance of a footloose industry due to almost negligible fixed costs, institutional investors are attracted to highly dynamic urban centres and on the 20 year time horizon, appear surprisingly sticky in their location preference. This is consistent with the belief from previous literature that institutional investors choose their urban locations due to the ubiquity of high quality information and access to advanced professional services commensurate to the investor's role as a command and control node in the broader economy. A close examination of the data concludes that while New York State may be in relative decline nationally, it is still the fastest growing location in absolute terms and remains firmly at the top of the US investor urban hierarchy. Lastly, the machine learning classification of investment strategies shows that California-based investors have unexpectedly poor performance across a broad range of identified strategies, however this can be explained by California's venture capital culture, hinting that local knowledge and culture can influence investment strategy decisions.

Published
2020

75. Conspicuous consumption: The figure of the serial killer as cannibal in the age of capitalism

Author

Lefebvre, Martin

Subjects
Social sciences
Abstract

A conceptual model for the way ideas, conceptions, or feelings are represented of figured in memory with the help of the imagination which constitutes part of the children's memory-image of the serial killer in both fact and fiction The rhetorical network that surrounds cannibalism offers imagination a topos for the memory-image of the serial killer.

Published
2005

77. ECL I/O buffers for BiCMOS integrated systems: a tutorial overview

Author

Pickles, Neil S. and Lefebvre, Martin C.

Subjects
BiCMOS -- Analysis, Emitter-coupled logic -- Analysis, I/O devices -- Analysis, Integrated circuits -- Analysis, Business, Education, Electronics, Electronics and electrical industries
Abstract

In this paper we describe the methodology for the design and layout of fast BiCMOS ECL I/O buffers. Principles of ECL circuit operation are described with emphasis on the NOR/OR gate and the bandgap voltage reference. A comparison of ECL 10K and 100K logic families is presented as well as complete designs for an input and output buffer. The pad macros are temperature- and supply-voltage-compensated and have nominal rise and fall times of 400 ps. Index Terms - Design of I/O buffers, digital logic devices, ECL buffers.

Published
1997

78. Learning without feedback: Direct random target projection as a feedback-alignment algorithm with layerwise feedforward training

Author

UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique, Frenkel, Charlotte, Lefebvre, Martin, Bol, David, UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique, Frenkel, Charlotte, Lefebvre, Martin, and Bol, David

Abstract

While the backpropagation of error algorithm allowed for a rapid rise in the development and deployment of artificial neural networks, two key issues currently preclude biological plausibility: (i) symmetry is required between forward and backward weights, which is known as the weight transport problem, and (ii) updates are locked before both the forward and backward passes have been completed. There is thus a growing interest in the development of training algorithms that release these constraints and ensure locality in both parameters and error signals while minimizing the training performance penalty. The feedback alignment (FA) algorithm uses fixed random feedback weights and shows that the network learns to align its forward and backward weights to maximize error gradient information. The direct feedback alignment (DFA) variation directly propagates the output error to each hidden layer through fixed random connectivity matrices. In this work, we show that using only the error sign is sufficient to maintain feedback alignment and to provide learning in the hidden layers. As in classification problems the error sign information is already contained in the targets (i.e. one-hot-encoded labels), using the latter as a proxy for the error brings three advantages: (i) it solves the weight transport problem by eliminating the requirement for an explicit feedback pathway, which also reduces the computational workload, (ii) it reduces memory requirements by removing update locking, allowing for weight updates to be computed in each layer independently without requiring a full forward pass, and (iii) it leads to a purely feedforward and low-cost algorithm that only requires a label-dependent random vector selection to estimate the layerwise loss gradients. Therefore, in this work, we propose the direct random target projection (DRTP) algorithm and demonstrate on the MNIST and CIFAR-10 datasets that, despite the absence of an explicit error feedback, DRTP can still lie cl

Published
2019

79. MAV payload: An air-quality monitoring system for integration inside a drone

Author

Andrikopoulou, Christina, Le Calvé, Stéphane, Sutter, Christophe, Noël, Florian, Rodríguez-Cuevas, Alberto, Lara-Ibeas, Irene, Kiefer, Renaud, Verdines, Mark, Pavot, Thomas, Lefebvre, Martin, Ettrich, Jörg, Schröder, Werner, Junk, Stefan, Schrock, Steffen, Institut de chimie et procédés pour l'énergie, l'environnement et la santé (ICPEES), Université de Strasbourg (UNISTRA)-Matériaux et nanosciences d'Alsace (FMNGE), Institut de Chimie du CNRS (INC)-Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)-Institut de Chimie du CNRS (INC)-Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS), In’Air Solutions [Strasbourg, France], Institut National des Sciences Appliquées - Strasbourg (INSA Strasbourg), Institut National des Sciences Appliquées (INSA), University of Applied Sciences Offenburg, University of Applied Sciences [Offenburg], European Project: ELCOD, Le Calvé, Stéphane, Endurance LOW COst Drone - ELCOD - INCOMING, Université de Strasbourg (UNISTRA)-Institut de Chimie du CNRS (INC)-Centre National de la Recherche Scientifique (CNRS)-Matériaux et Nanosciences Grand-Est (MNGE), and Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Institut National de la Santé et de la Recherche Médicale (INSERM)-Institut de Chimie du CNRS (INC)-Centre National de la Recherche Scientifique (CNRS)-Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)

Subjects
[SDE] Environmental Sciences, [SPI]Engineering Sciences [physics], [CHIM.ANAL] Chemical Sciences/Analytical chemistry, [SPI] Engineering Sciences [physics], [CHIM.ANAL]Chemical Sciences/Analytical chemistry, [SDE]Environmental Sciences, ComputingMilieux_MISCELLANEOUS
Abstract

International audience

Published
2018

80. Carbon fibre/pvc foam sandwich composite modelization for MAVs & long range drones structures

Author

Lefebvre, Martin, Kiefer, Renaud, Vedrines, Marc, Chibane, Hicham, univOAK, Archive ouverte, Laboratoire des sciences de l'ingénieur, de l'informatique et de l'imagerie (ICube), École Nationale du Génie de l'Eau et de l'Environnement de Strasbourg (ENGEES)-Université de Strasbourg (UNISTRA)-Institut National des Sciences Appliquées - Strasbourg (INSA Strasbourg), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Les Hôpitaux Universitaires de Strasbourg (HUS)-Centre National de la Recherche Scientifique (CNRS)-Matériaux et Nanosciences Grand-Est (MNGE), Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Institut National de la Santé et de la Recherche Médicale (INSERM)-Institut de Chimie du CNRS (INC)-Centre National de la Recherche Scientifique (CNRS)-Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Institut National de la Santé et de la Recherche Médicale (INSERM)-Institut de Chimie du CNRS (INC)-Centre National de la Recherche Scientifique (CNRS)-Réseau nanophotonique et optique, Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Centre National de la Recherche Scientifique (CNRS)-Université de Strasbourg (UNISTRA)-Centre National de la Recherche Scientifique (CNRS), Institut National des Sciences Appliquées - Strasbourg (INSA Strasbourg), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université de Strasbourg (UNISTRA)-Centre National de la Recherche Scientifique (CNRS)-École Nationale du Génie de l'Eau et de l'Environnement de Strasbourg (ENGEES)-Réseau nanophotonique et optique, Centre National de la Recherche Scientifique (CNRS)-Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Centre National de la Recherche Scientifique (CNRS)-Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Matériaux et nanosciences d'Alsace (FMNGE), and Institut de Chimie du CNRS (INC)-Université de Strasbourg (UNISTRA)-Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA))-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)-Institut de Chimie du CNRS (INC)-Université de Strasbourg (UNISTRA)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)

Subjects
[SPI.AUTO] Engineering Sciences [physics]/Automatic, [SPI.MECA.MSMECA] Engineering Sciences [physics]/Mechanics [physics.med-ph]/Materials and structures in mechanics [physics.class-ph], [SPI.MECA.MSMECA]Engineering Sciences [physics]/Mechanics [physics.med-ph]/Materials and structures in mechanics [physics.class-ph], [SPI.AUTO]Engineering Sciences [physics]/Automatic
Abstract

International audience; This paper deals with the characterization, modelization and simulation of a composite structure used for MAV or generally for drones design. Composite structures are extremely difficult to simulate due to the anisotropic behaviour. The first part of the article is focused on carbon fiber/PVC foam (AIREX) sandwich composite characterization with the design of experiments method on tensile tests. This method gives equations, which describe the material mechanical behaviour (Young’s modulus, tensile strength) depending on factors values. The second part of the article deals with the improvement of mechanical simulation of an anisotropic material with the goal to get an accurate model and to generalised properties to an entire structure. Then, the macroscopic mechanical properties of the most performing sandwich will be obtained with the global composite behaviour matrix 6x6, in order to be integrated into a Finite Element Analysis software (CREO / Simulate) to simulate a fixed wing behaviour. Finally, comparisons between experiment and numerical simulation on the wing will give promising results, and simulations will reveal substantial differences between tension & compression in a flexural solicitation.

Published
2018

82. Design, modelling and prototyping of an electrodynamic self-bearing axial-flux permanent magnet motor

Author

Lefebvre, Martin, UCL - Ecole polytechnique de Louvain, and Dehez, Bruno

Subjects
Axial-flux, Magnet, Motor, Self-bearing, Synchronous motor, Bearing, Electrodynamic
Abstract

Two topics have been the subject of numerous researches in the field of rotating machineries those past years. On the one hand, electrodynamic magnetic bearings (EDBs) ensure a contactless guidance of the rotor relying on the interaction between the permanent magnets (PMs) magnetic fields and the currents they induce in the conducting parts of the bearing structure. On the other hand, axial-flux permanent magnet (AFPM) machines allow to drive a PMs rotor disk by injecting three-phased currents in the windings. Topologies of electrodynamic thrust bearings (EDTBs) and AFPM machines present a significant number of similarities. Starting from this observation, this Master's thesis introduces a new machine topology combining both a bearing and motor behavior to obtain a self-bearing AFPM machine in a single winding. First, a model describing the axial and rotational dynamics of such a machine is developed. This model directly relates the axial displacement to the force developed by the EDTB and the rotational motion to the driving torque generated by the AFPM machine and the drag torque arising from the EDTB. The obtained model is constituted of eight non-linear first-order differential equations with a set of twelve parameters. The calculations required to determine those parameters using static FEM simulations are also detailed. Furthermore, this model can be used to obtain several results regarding the quasi-static and dynamic behavior of the self-bearing AFPM machine. Based on the linearization of the obtained model, this work assesses the stability properties of the system under constant rotational speed, based on a root locus analysis. The evolution of the quasi-static equilibrium points as a function of the spin speed is also discussed for a fixed axial displacement or a given external force applied to the rotor. Moreover, this model allows to simultaneously obtain the axial and rotational dynamic behavior of the self-bearing AFPM machine in various situations, e.g. at start-up and when subjected to external force perturbations or variations of the reference spin speed. A last use of the model is to determine optimal dimensions and parameters for the machine, based on a set of criteria describing the EDTB and AFPM motor performance. In this work, this purpose is achieved through a parametric analysis, but this model can be used to conduct a conventional optimization on the machine parameters. Finally, the prototype of a self-bearing AFPM motor and the associated testbench are developed as a proof-of-concept. The functioning of the EDTB and AFPM motor alone has been partially validated based on experimental measurements, but the fact that the prototype is not completely functional has kept us from determining the performance of the complete system. However, critical issues in the development of such a machine have been highlighted and solutions to avoid and solve them have been proposed. This Master's thesis thus paves the way for further works combining a fully passive magnetic guidance and a motor behavior within the same device. Master [120] : ingénieur civil électromécanicien, Université catholique de Louvain, 2017

Published
2017

83. Dynamic modelling of passive electrodynamic self-bearing axial-flux permanent magnet machines

Author

Lefebvre, Martin, Van Verdeghem, Joachim, Kluyskens, Virginie, Dehez, Bruno, IEEE International Electric Machines and Drives Conference (IEMDC), and UCL - SST/IMMC/MEED - Mechatronic, Electrical Energy, and Dynamics Systems

Subjects
Engineering, Self-Bearing Motors, Mechanical engineering, 02 engineering and technology, Passive, 01 natural sciences, Dynamic model, law.invention, Control theory, law, 0103 physical sciences, 0202 electrical engineering, electronic engineering, information engineering, Torque, Thrust, 010302 applied physics, Bearing (mechanical), business.industry, 020208 electrical & electronic engineering, Rotation around a fixed axis, Root locus, Rotational speed, Electrodynamic, Thrust bearing, Magnet, Bearing, Synchronous motor, business
Abstract

Topologies of electrodynamic thrust bearings (EDTB) and axial flux permanent magnet (AFPM) machines present a significant number of similarities. Starting from this observation, the present paper introduces a new machine topology combining both a bearing and motor behavior to obtain a self-bearing AFPM machine in a single winding. In addition, this paper presents a model describing the axial and rotational dynamics of such a machine. This model directly relates the axial displacement to the force developed by the EDTB and the rotational motion to the driving torque generated by the AFPM machine and the drag torque arising from the EDTB. The calculations required to determine the model parameters are also detailed. Furthermore, this work assesses the stability properties of the system under constant rotational speed, based on a root locus analysis. Finally, a practical case is considered to assess the performance of the system.

Published
2017

85. Dynamic modelling of passive electrodynamic self-bearing axial-flux permanent magnet machines

Author

UCL - SST/IMMC/MEED - Mechatronic, Electrical Energy, and Dynamics Systems, Lefebvre, Martin, Van Verdeghem, Joachim, Kluyskens, Virginie, Dehez, Bruno, IEEE International Electric Machines and Drives Conference (IEMDC), UCL - SST/IMMC/MEED - Mechatronic, Electrical Energy, and Dynamics Systems, Lefebvre, Martin, Van Verdeghem, Joachim, Kluyskens, Virginie, Dehez, Bruno, and IEEE International Electric Machines and Drives Conference (IEMDC)

Abstract

Topologies of electrodynamic thrust bearings (EDTB) and axial flux permanent magnet (AFPM) machines present a significant number of similarities. Starting from this observation, the present paper introduces a new machine topology combining both a bearing and motor behavior to obtain a self-bearing AFPM machine in a single winding. In addition, this paper presents a model describing the axial and rotational dynamics of such a machine. This model directly relates the axial displacement to the force developed by the EDTB and the rotational motion to the driving torque generated by the AFPM machine and the drag torque arising from the EDTB. The calculations required to determine the model parameters are also detailed. Furthermore, this work assesses the stability properties of the system under constant rotational speed, based on a root locus analysis. Finally, a practical case is considered to assess the performance of the system.

Published
2017

86. Les chiens de faïence. Introduction à Chapelles et querelles des théories du cinéma

Author

Jullier, Laurent, Lefebvre, Martin, IRCAV - Institut de Recherche sur le Cinéma et l'Audiovisuel - EA 185 (IRCAV), Université Sorbonne Nouvelle - Paris 3-LABEX ICCA, Université Paris 13 (UP13)-Université Sorbonne Nouvelle - Paris 3-Centre National de la Recherche Scientifique (CNRS)-Université de Paris (UP)-Université Sorbonne Paris Nord-Université Paris 13 (UP13)-Centre National de la Recherche Scientifique (CNRS)-Université de Paris (UP)-Université Sorbonne Paris Nord, and Concordia University [Montreal]

Subjects
[SHS.INFO]Humanities and Social Sciences/Library and information sciences
Published
2016

87. A Discussion on Gender Studies

Author

Beugnet, Martine, Cervulle, Maxime, Coulomb-Gully, Marlène, Sellier, Geneviève, Lefebvre, Martin, Jullier, Laurent, Laboratoire de Recherche sur les Cultures Anglophones (LARCA UMR 8225), Université Paris Diderot - Paris 7 (UPD7)-Centre National de la Recherche Scientifique (CNRS), Centre d'études sur les médias, les technologies et l'internationalisation (CEMTI), Université Paris 8 Vincennes-Saint-Denis (UP8), Laboratoire d'Etudes et de Recherches Appliquées en Sciences Sociales (LERASS), Université Toulouse - Jean Jaurès (UT2J)-Université Toulouse III - Paul Sabatier (UT3), Université Fédérale Toulouse Midi-Pyrénées-Université Fédérale Toulouse Midi-Pyrénées-Université Paul-Valéry - Montpellier 3 (UPVM), Cultures, Littératures, Arts, Représentations, Esthétiques (CLARE), Université Bordeaux Montaigne, Institut National des Sciences Appliquées - Strasbourg (INSA Strasbourg), Institut National des Sciences Appliquées (INSA), Laboratoire Lorrain de Sciences Sociales (2L2S), Université de Lorraine (UL), Université Paul-Valéry - Montpellier 3 (UPVM)-Université Toulouse - Jean Jaurès (UT2J)-Université Toulouse III - Paul Sabatier (UT3), and Université Fédérale Toulouse Midi-Pyrénées-Université Fédérale Toulouse Midi-Pyrénées

Subjects
gender studies, epistemology and research in the Humanities, théories du cinéma, [SHS.INFO]Humanities and Social Sciences/Library and information sciences, Film theory, queer studies, épistémologie de la recherche en sciences humaines, [SHS]Humanities and Social Sciences
Abstract

International audience; A debate where scholars address research methodology as well as heuristic and epistemic concerns relevant to Gender Studies in as much as as they apply to film and moving image studies.; Débat entre différents chercheurs abordant des questions relatives à la méthodologie, aux capacités heuristiques et au bien-fondé épistémologique des Gender Studies appliquées au cinéma et à l’audiovisuel.

Published
2016
Full Text
View/download PDF

89. Présentation

Author

Gaudreault, André, Lefebvre, Martin, Technès (International Research Partnership on Cinema Technology), Université de Montréal. Faculté des arts et des sciences. Département d'histoire de l'art et d'études cinématographiques, and Technès (Partenariat international de recherche sur les techniques et technologies du cinéma)

Published
2015

90. A 0.086-mm$^2$ 12.7-pJ/SOP 64k-Synapse 256-Neuron Online-Learning Digital Spiking Neuromorphic Processor in 28-nm CMOS.

Author

Frenkel, Charlotte, Lefebvre, Martin, Legat, Jean-Didier, and Bol, David

Abstract

Shifting computing architectures from von Neumann to event-based spiking neural networks (SNNs) uncovers new opportunities for low-power processing of sensory data in applications such as vision or sensorimotor control. Exploring roads toward cognitive SNNs requires the design of compact, low-power and versatile experimentation platforms with the key requirement of online learning in order to adapt and learn new features in uncontrolled environments. However, embedding online learning in SNNs is currently hindered by high incurred complexity and area overheads. In this paper, we present ODIN, a 0.086-mm $^2$ 64k-synapse 256-neuron online-learning digital spiking neuromorphic processor in 28-nm FDSOI CMOS achieving a minimum energy per synaptic operation (SOP) of 12.7 pJ. It leverages an efficient implementation of the spike-driven synaptic plasticity (SDSP) learning rule for high-density embedded online learning with only 0.68  $\mu$ m $^2$ per 4-bit synapse. Neurons can be independently configured as a standard leaky integrate-and-fire model or as a custom phenomenological model that emulates the 20 Izhikevich behaviors found in biological spiking neurons. Using a single presentation of 6k 16 × 16 MNIST training images to a single-layer fully-connected 10-neuron network with on-chip SDSP-based learning, ODIN achieves a classification accuracy of 84.5%, while consuming only 15 nJ/inference at 0.55 V using rank order coding. ODIN thus enables further developments toward cognitive neuromorphic devices for low-power, adaptive and low-cost processing. [ABSTRACT FROM AUTHOR]

Published
2019
Full Text
View/download PDF

91. A Spatial Exploration of Institutional Investment in Canada for the Year 2010

Author

Lefebvre, Martin R

Subjects
Urban Studies, Canada, Financial Geography, Institutional Investment, Economic Geography, Other Geography, Human Geography, Political Economy, Finance
Abstract

Using measures of central tendency, the average nonUS-based institutional investor has more capital invested in securities than his US-based counterpart. The present study shows that US-based investors favour manufacturing companies, whereas Canadian investors prefer companies based in natural resources. Nationally, Toronto acts more as the centre of gravity for Canadian institutional investors than New York City does for the United States. Comparatively, Toronto accounts for 70% of all Canadian investors while New York accounts for only 30% of the American total, despite it being the city with the most capital invested worldwide. Notwithstanding Alberta’s oil boom, inter-provincial investment capital show that Ontario is the leading province for receiving investment capital from Canadian sources, as well as investing institutional capital across Canada. Locally, investors in Montreal and Toronto are more spatially clustered than their American counterparts (Chicago and New York), since both Canadian cities are devoid of investors located in suburban areas.

Published
2014

96. Le plan à la première personne. Technologie et subjectivité dans le paysage postcinematographique

Author

Gaudreault, André, Lefebvre, Martin, Eugeni, Ruggero, Eugeni, Ruggero (ORCID:0000-0002-7198-1606), Gaudreault, André, Lefebvre, Martin, Eugeni, Ruggero, and Eugeni, Ruggero (ORCID:0000-0002-7198-1606)

Abstract

The article focus on the new forms of the subjective image in audiovisual media, defining the "first person shot" as a new kind of post-media semiotic figure. The first-person shot appears on different occasions throughout different media, from combat videos recorded with helmet cams and distributed via the Internet to video-surveillance footage re-used in art installations, and from video clips of live events taken with video cell phones and broadcasted by television news to first-person video games, to cite just a few examples. The core of my paper is devoted to reconstructing the first-person shot’s technological and stylistic genealogy. I argue that although firmly a part of contemporary cinema, it constitutes a radically intermedial and post-cinematographic technique, since it stems from a network of “remediations” and “deep-remixes” involving different media agencies and players – from mainstream cinema to independent television and video producers, from the video-game industry to military and surveillance uses of audiovisuals. In my conclusions, I rapidly sketch two questions related to the first-person shot. First, I ask whether the first-person shot should or not be considered not just as an audio-visual technique but also as a well-defined semiotic figure. My answer is positive; moreover, I argue that a formal assessment of the first-person shot entails a new systematization of point-of-view issue within the semiotics of audiovisual media. The second question concerns the cultural foundation of the first-person shot. In this regard, I propose to consider it as a kind of “symbolic form”, implying a specific conception of the subject as a hybrid, dynamical and relational entity – an idea of the subject and sub-jectivity chiming with contemporary conceptions of phenomenological neuroscience.

Published
2015

Catalog

Books, media, physical & digital resources
See catalog results