HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
11652 research outputs found
Sort by
Modelling a simplified device—a wind musical instrument—as an educational tool to study the behavior of a one-dimensional resonator
No full textInternational audienceThe aim of this paper is to show how the analytical modelling of a simplified acoustic device (herein a simplified wind musical instrument) may be a relevant educational way to study several aspects of the behavior of an idealized component [herein a one-dimensional (1-D) resonator], which is the core component of the device considered. A time-dependent source, whether sinusoidal, impulse, or representing approximate valve-effects of the reed, is coupled to the resonator in the analytical modelling. The unavoidable thermo-viscous parietal dissipation, as well as an approximate radiation effect, are accounted for. After a brief presentation of the problem, precisely posed and solved in detail analytically in an approximate manner, results in both the frequency domain and in the time domain are presented. These results include the properties of the transient behaviors (both the starting transient and the transient decay), illustrating the main features of the resonator.</jats:p
Analyse comparative de la modération des réseaux sociaux numériques. Etude de cas sur la DSA Transparency Database.
No full textIn the last decade, many scholars have worked on content moderation from different perspectives. Many have focused on the issues related to censorship and freedom of expression from a law perspective (focusing on the regulation framework i.e.) ; others have tried to work with digital methods either to work with great corpus of data, either to try to understand how algorithmic content moderation actually works on each platform; etc. This study aims at comparing how content moderation works on different social media. Working with data from the DSA Transparency Database, we aim at comparing how platforms like Facebook, Instagram, Twitter/X, TikTok (etc.) moderate their content online. Therefore, the study explores different questions: What type of content is more likely to be deleted?What is the proportional difference between content being deleted by humans and machines?How do the proportions of content being deleted differ across different social media platforms?How do the proportions of content deleted by humans and machines differ across social media platforms?In which category do we find the highest partial/temporary suspension incidence
Position Estimation using the Radical Axis Gauss Newton Algorithm: Experimental Analysis
No full textInternational audienceGenerally a UAV is mounted with multiple sensors to help in their navigation including altitude and position sensors. Although there are currently different sensors to determine the flight altitude, the acquisition of the position at every moment is a hard task to be achieved mainly in crowded scenarios where the multiple obstacles make difficult the transmission of signal coming from satellites (for global navigation systems) or base stations which do the functions of anchors in some localization systems. Another issue of this kind of scenarios is the low probability to obtain redundancy systems due to obstruction of the signal. For the case where there is only the minimum information required to estimate the position, in this work the RA-GN algorithm proposed before is applied to a localization system. In order to prove the achieved efficiency by the RA-GN algorithm as well as to analyze its accuracy, a measurement campaign was conducted in a semi-forest environment. Results reported here show that the RA-GN algorithm is able to improve the accuracy of estimation in real conditions into crowded environments, specially under critical situations where the signal is perturbed by the obstacles and even by the orientation of the tag respect to the achors compared with other positioning methods. I
Traiter les Incohérences dans les Préférences (numériques) à l'aide de la Théorie des Possibilités
No full textNational audienceGathering the preferences of a user in order to make correct recommendations becomes a difficult task when said user is uncertain of her choices. Using possibility theory as a means of modelling and detecting this uncertainty, we propose in this paper methods based on information fusion to make inferences despite observed inconsistencies due to user errors. While our approach is general, we illustrate its potential benefits on synthetic experiments using weighted averages as preference models.Recueillir les préférences d'une utilisatrice afin de faire des recommandations correctes est une tâche difficile dès lors que cette utilisatrice est incertaine de ses choix. En utilisant la théorie des possibilités comme moyen de modéliser et de détecter cette incertitude, nous proposons dans cet article des méthodes basées sur la fusion d'informations dans le but d'inférer malgré les incohérences observées dues aux erreurs de l'utilisatrice. Bien que notre approche soit générale, nous illustrons ses avantages potentiels sur des expériences synthétiques avec des moyennes pondérées comme modèle de préférence
Reconfigurable Manufacturing Systems, Layout, Process Planning, Scheduling
No full textInternational audienc
Décision prudente pour les ensembles d'arbres
No full textInternational audienceCautious classifiers are designed to make indeterminate decisions when the uncertainty on the input data or the model output is too high, so as to reduce the risk of making wrong decisions. In this paper, we propose two cautious decision-making procedures, by aggregating trees providing probability intervals constructed via the imprecise Dirichlet model. The trees are aggregated in the belief functions framework, by maximizing the lower expected discounted utility, so as to achieve a good compromise between model accuracy and determinacy. They can be regarded as generalizations of the two classical aggregation strategies for tree ensembles, i.e., averaging and voting. The efficiency and performance of the proposed procedures are tested on random forests and illustrated on three UCI datasets.Les classificateurs prudents sont conçus pour prendre des décisions indéterminées lorsque l'incertitude sur les données d'entrée ou la sortie du modèle est trop élevée, afin de réduire le risque de prendre de mauvaises décisions. Dans cet article, nous proposons deux procédures de prise de décision prudente, en agrégeant des arbres fournissant des intervalles de probabilité construits via le modèle de Dirichlet imprécis. Les arbres sont agrégés dans le cadre des fonctions de croyance, en maximisant l'utilité escomptée actualisée la plus faible, afin de parvenir à un bon compromis entre la précision du modèle et le caractère déterminant. Elles peuvent être considérées comme des généralisations des deux stratégies d'agrégation classiques pour les ensembles d'arbres, à savoir le calcul de la moyenne et le vote. L'efficacité et la performance des procédures proposées sont testées sur des forêts aléatoires et illustrées sur trois ensembles de données de l'UCI
Comment les Nouvelles Routes de la Soie (BRI) changent-elles les formes d'urbanisation en Asie du sud-est?
No full textInternational audienceThis paper examines how Chinese transnational investments, as (re)framed in the Belt and Road Initiative (BRI), contribute to changes in urbanisation processes in Southeast Asia. On the ground, the BRI becomes contextualised and intersects with local and national development trajectories. The growing presence of Chinese actors in the region intensifies urban dynamics, especially in secondary cities and emerging urban sites, where the BRI is used as a lever for local internationalisation strategies. The heterogeneous nature of the links between the BRI and various large urban projects is demonstrated on the basis of case studies involving changing consortia of private and public Chinese and Southeast Asian actors. A regional approach allows us to identify connections and shared processes across Southeast Asian countries. It provides a historically grounded understanding of how the BRI incorporates long‐term interactions with China and more recent partnerships in Southeast Asian countries. The paper paves the way for a research agenda that contests the image of China as a monolithic actor implementing the BRI uniformly and consistently. Further analyses are needed to examine systems and networks of actors as well as the local urban politics that affect the BRI on the ground
Un modèle d’édition numérique pour le corpus d’Elisa Chimenti : réflexions préliminaires
No full textInternational audienc
An Overview of Decision-Making in Autonomous Vehicles
No full textInternational audienceIn this review we have surveyed the literature on the autonomous vehicle focusing on the decision-making architecture. The main contribution of this paper is to provide an upto-date reference that tries to cover as much as possible this broad topic rapidly. Therefore, after analyzing the body of literature, we were able to present a detailed overview of the usual anatomy of the decision-making system, review the coverage of public reviews and surveys available around the topic, present the environment and vehicle representation, and showcase the literature through: trajectory planning, risk-uncertainty assessment, and mimicking human behaviour mainly using perception-based methods. We also classified the literature according to scenarios covered, action-space used, and whether or not they emphasize the time horizon or frequency or time-step
