HAL Portal de Univ. Gustave Eiffel
Not a member yet
72847 research outputs found
Sort by
Détection automatisée de véhicules lents sur fil de droite en milieu autoroutier dans un environnement partiellement composé de véhicules connectés
No full textInternational audienceWith the expanding development of C-ITS services and their field implementation, our driving experience is now occurring under a partially connected environment. Whether through embedded smartphones or onboard equipment, vehicles are getting connected and regularly emit high-frequency safety messages (CAM in Europe or BSM in the USA) regarding their status. In this paper, as an alternative to the usual methods that rely on expensive dedicated cameras, we explore the potential of passive data resources to feed a slow obstacle detection process performed in near-real time. Contrary to the recent literature focusing on the development of dynamic obstacle detection to expand autonomous skills through expensive dedicated sensors, we adopt the road managers’ perspective. We assume the existence of a monitoring and management center, potentially decentralized to Road-Side Units, collecting the data stream continuously, analyzing it, and enabling it to broadcast safety warning messages to connected vehicles located immediately upstream of the identified slow obstacle. The two-step methodology is based on (i) an automatic lane change detection process followed by (ii) a weighting process and statistical analysis of the space–time scatter plots generated by detected lane changes over a sliding time window. Simulation-based results highlight that despite a low share of connected vehicles, a stationary obstacle can be detected on average at 3 min, while longer delays (5 min) are required when the obstacle is moving between 30 km/h and 50 km/h. Furthermore, disseminating warning messages upstream can improve safety and traffic performance by up to 10 percent for low traffic conditions
Moins de règles pour plus de haies ? Enquête sur un dispositif de simplification administrative
No full textInternational audienc
Présence Integral Presence at Work as a means to Spiritually Transform Organizations
No full textInternational audienc
De la RSE à la Responsabilité Territoriale des Entreprises : une exploration anthropologique en Polynésie Française
No full textInternational audienceEn Polynésie française, la Responsabilité Sociale des Entreprises (RSE) reste embryonnaire mais suscite un intérêt croissant parmi les organisations locales. Cette recherche explore leur contribution à l’amélioration des pratiques écoresponsables dans un contexte insulaire. S’appuyant sur une ethnographie menée en 2023 incluant 63 entretiens et des observations participantes dans trois entreprises, les résultats révèlent des initiatives ascendantes : gestion des déchets, sensibilisation environnementale et partenariats communautaires même si ces dynamiques se heurtent parfois à des tensions culturelles entre les stratégies organisationnelles et les comportements individuels. En dépassant une application de normes globales, la RSE peut s’inscrire localement dans une gouvernance adaptée aux réalités territoriales ouvrant à une conceptualisation d’une Responsabilité Territoriale des Entreprises (RTE), articulant acteurs locaux et transnationaux
Low-Complexity Approach to Intelligent SHM by Combining Machine Learning Models Using Single-Sensor Data
No full textInternational audienceStructural Health Monitoring (SHM) systems, when applied to large civil engineering structures such as bridges, process high-volume data and run computationally intensive algorithms, which typically require important processing power to ensure low inference latency to enable real-time monitoring and rapid decision-making. In this study, we propose a novel resource-efficient approach to optimizing SHM for civil structures. The methodology integrates lightweight machine learning models that rely exclusively on single-sensor data, enabling deployment at the sensor level (smart sensors). By aggregating outputs from multiple sensors, the approach captures spatial information, introduces diversity, and benefits from an averaging effect, significantly improving overall performance compared to individual sensor-based predictions. This single-sensor strategy ensures low computational complexity while maintaining high accuracy, making it particularly suitable for resource-constrained environments. To evaluate the effectiveness of the proposed methodology, we applied it to the Z24 benchmark dataset, a widely recognized SHM resource for civil structures. The objective was to classify various damage scenarios based on data collected from accelerometers deployed on the bridge. The results demonstrate competitive performance with minimal computational complexity, highlighting the scalability and suitability of such an approach for large-scale SHM applications. Ultimately, this study underscores the potential of resource-efficient SHM solutions, contributing to developing sustainable and intelligent monitoring systems.</div
Segmenting France Across Four Centuries
No full textInternational audienceHistorical maps offer an invaluable perspective into territory evolution across past centuries, long before satellite or remote sensing technologies existed. Deep learning methods have shown promising results in segmenting historical maps, but publicly available datasets typically focus on a single map type or period, require extensive and costly annotations, and are not suited for nationwide, long-term analyses. In this paper, we introduce a new dataset of historical maps tailored for analyzing large-scale, long-term land use and land cover evolution with limited annotations. Spanning metropolitan France (548,305 km^2), our dataset contains three map collections from the 18th, 19th, and 20th centuries. We provide both comprehensive modern labels and 22,878 km^2 of manually annotated historical labels for the 18th and 19th century maps. Our dataset illustrates the complexity of the segmentation task, featuring stylistic inconsistencies, interpretive ambiguities, and significant landscape changes (e.g., marshlands disappearing in favor of forests). We assess the difficulty of these challenges by benchmarking three approaches: a fully-supervised model trained with historical labels, and two weakly-supervised models that rely only on modern annotations. The latter either use the modern labels directly or first perform image-to-image translation to address the stylistic gap between historical and contemporary maps. Finally, we discuss how these methods can support long-term environment monitoring, offering insights into centuries of landscape transformation. Our repository is publicly available on GitHub
Assessment of genetically modified soybean MON 87708 for renewal authorisation under Regulation (EC) No 1829/2003 (dossier GMFF‐2023‐21237)
No full textFollowing the submission of dossier GMFF‐2023‐21237 under Regulation (EC) No 1829/2003 from Bayer CropScience LP, the Panel on Genetically Modified Organisms of the European Food Safety Authority was asked to deliver a scientific risk assessment on the data submitted in the context of the renewal of authorisation application for the herbicide tolerant genetically modified soybean MON 87708, for food and feed uses, excluding cultivation within the European Union. The data received in the context of this renewal application contained post‐market environmental monitoring reports, an evaluation of the literature retrieved by a scoping review, a search for additional studies performed by or on behalf of the applicant and updated bioinformatics analyses. The GMO Panel assessed these data for possible new hazards, modified exposure or new scientific uncertainties identified during the authorisation period and not previously assessed in the context of the original application. Under the assumption that the DNA sequence of the event in soybean MON 87708 considered for renewal is identical to the sequence of the originally assessed event, the GMO Panel concludes that there is no evidence in renewal dossier GMFF‐2023‐21237 for new hazards, modified exposure or scientific uncertainties that would change the conclusions of the original risk assessment on soybean MON 87708
Finite strain micro-poro-mechanics: Formulation and compared analysis with macro-poro-mechanics
No full textInternational audiencePorous materials are ubiquitous in nature -notably living tissues, which often undergo large deformations and engineering applications. Poromechanics is an established theory to model the response of such materials; however, it is limited in its description of microscale phenomena, and structure-properties relationships. In this paper, we propose a microscopic poromechanical model based on a novel formulation of the micro-poro-mechanics problem, which allows to compute the response of any porous periodic microstructure to any loading involving fluid pressure, macroscopic strain, and/or macroscopic stress. We systematically compare the global response of our micro-model to macro-poromechanics, in both the infinitesimal and finite strain settings, and investigate in particular three mechanisms, namely solid compressibility, strain-pressure coupling and deviatoric-volumetric strain coupling. We notably illustrate how the micro-model can be used to derive macroscopic parameters, and how these parameters depend on microscopic features like pore shape, porosity, material properties, etc. This modeling framework will be the basis for powerful micro-poro-mechanical models of various materials and tissues, where pore-scale phenomena can be incorporated explicitly
Explorer la relation entre la cyclabilité perçue et l'usage inclusif de la micro-mobilité en fonction du genre. Approche comparative dans 53 villes françaises
No full textInternational audienceAs the utilization of micromobility continues to experience growth and diversification, while simultaneously gaining recognition as an environmentally-friendly mode of transportation, it remains predominantly male-dominated. Recent scientific literature has highlighted the importance of inclusive strategies, demonstrating a strong correlation between the gender gap in cycling participation and the overall cycling levels within a given area. This indirect relationship necessitates identifying the factors that promote gender-inclusive bicycle usage. Focusing on the French context, the key objectives of this empirical research are (i) measuring gender inequalities in the use of bike and emerging micromobility at the municipal level, (ii) assessing the influence of built environment and urban design on the gendered modal share of cyclists, and (iii) comparing and clustering the investigated French cities with the development of an index that takes into account gender equity, the modal share of cycling, and the perceived bikeability. By drawing from two distinct databases based on the use of micromobility and the subjective bikeability of cities and by conducting quantitative observations, this original study sheds light on the significant connection between gender-balanced cycling distribution, cycling modal share, cycling infrastructure presence and perceived bikeability. This paper concludes that encouraging women to embrace cycling is not solely dependent on achieving a critical mass of cyclists or building cycling lanes. Instead, it emphasizes the need for the development of a comprehensive ’bicycle system’ that takes into account all aspects of bikeability. This innovative outcome leads to the categorization of examined cities based on the development of a gender-inclusive with cycling quality index. This exploration underscores the vital role of urban planning and offers recommendations for stakeholders regarding future policy initiatives.Alors que l'usage du vélo et de la micro-mobilité continuent de connaître une croissance et une diversification, tout en gagnant en reconnaissance en tant que modes de déplacement respectueux de l'environnement, cette mobilité individuelle légère reste majoritairement dominée par les usagers masculins. La littérature scientifique récente a souligné l'importance de stratégies englobant l'ensemble de la population, révélant une corrélation significative entre la pratique du vélo en fonction du genre et la part modale de ce mode dans un territoire donné. Cette relation indirecte nécessite d'identifier les facteurs favorisant la parité dans l'usage du vélo. Se concentrant sur le contexte français, les principaux objectifs de cette recherche empirique sont (i) de mesurer les inégalités de genre dans l'utilisation du vélo et de la micro-mobilité émergente au niveau municipal, (ii) d'évaluer l'influence de l'environnement urbain sur la répartition genrée du vélo et (iii) de comparer et de regrouper les villes françaises étudiées avec le développement d'un indice prenant en compte l'équité de genre, la part modale du vélo et la cyclabilité perçue. En s'appuyant sur deux bases de données distinctes basées sur l'usage du vélo et la cyclabilité subjective des villes, et en conduisant des observations quantitatives, cette étude originale met en lumière le lien significatif entre la distribution équilibrée des cyclistes en fonction du genre, la part modale du vélo, la présence d'infrastructures cyclables et la cyclabilité perçue. Cette publication conclut que la participation féminine au vélo et à la micro-mobilité ne dépend pas seulement de l'atteinte d'une masse critique de cyclistes ou de l'aménagement d'itinéraires cyclables. Au lieu de cela, elle souligne la nécessité de développer un « système vélo » qui prend en compte tous les aspects de la cyclabilité. Cette approche a dès lors mené à la catégorisation des villes examinées en fonction du développement d'un indice de qualité cyclable en lien avec la parité. Cette exploration souligne le rôle crucial de l'action de l'urbanisme et offre des recommandations aux acteurs de la fabrique urbaine