HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
11652 research outputs found
Sort by
High Integrity Localization with Bayesian Optimization for Information Filter Tuning with Fault Detection
No full textInternational audienceSafe navigation of autonomous vehicles relies on high integrity localization system based on a state estimation method, which strongly depends on the choice of parameters, such as the measurement noise covariance matrix. To avoid filter divergence, a fault detection and exclusion (FDE) procedure is also required. Selecting appropriate thresholds for the FDE step is challenging and affects the accuracy and the integrity of localization. This paper presents a multi-sensor data fusion based on the Information Filter (IF) with an auto-tuning method that relies on Bayesian Optimization (BO). The proposed method aims to optimize the measurement covariance matrix and the FDE thresholds. BO is well suited for non-convex and stochastic cost functions with local minima. A novel objective function is also designed to improve the accuracy and guarantee the integrity of the estimates, by ensuring consistent uncertainty regions. The objective function consists of two terms: one related to the error and the other to the quantile. The proposed approach is evaluated with experimental data from a vehicle equipped with wheel speed sensors, fused with Global Navigation Satellite System (GNSS) pseudorange measurements
Monitoring and Sensing System for People's Behavior During Fall Events Based on Mobility Analysis
No full textInternational audienceObjectivesObserving the activities of the elderly in natural life is a crucial issue nowadays to better understand their potential behavioral changes and predict risks. To this end, a comprehensive hardware and software infrastructure has been designed by a multidisciplinary team of researchers and pre-tested in a smart flat lab. It enables to collect relevant data and develop algorithms to analyze activities and detect changes such as falls, wandering or other risky situations. This study was carried out in a shared house by 12 independent elderly people. The study focuses on episodes of falls in the house, and analyzes mobility behavior before and after falls to observe the person's rehabilitation in the home.Materials and MethodsEach resident's room and the two shared spaces were equipped with motion and magnetic contact sensors to record movements and entry/exit activities. 9 months of data were collected and analyzed, highlighting patterns of activity and changes in these behaviors, particularly when a fall occurred and then when the usual behavior returned, if at all. Two levels of analysis were implemented: the detection of deviation in activity indicators for each individual, and the detection of drift in the established behavior pattern over time. The classification technique used to extract the patterns is the K-means partitioning algorithm. We also used the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) method to validate the performance of the K-means method.ResultsData analysis was carried out on the 4 falls recorded during the observation period, involving 4 of the house's occupants. The results highlight the relationship between model conduct and events related to falls and returns from hospitalization. Detection was validated by share house carers' annotations, acting as a ground truth, on the days when falls occurred. The first results of pattern recognition with clustering methods show that the K-means method provides more convincing results than the DBSCAN method. In this study, by observing the movement signals of residents who fell during the course of the study, we were able to identify characteristic post-fall behaviors
Using LLMs for post-OCR correction on historical French texts: A case study using synthetic data
No full textInternational audienceIn the wake of generative AI, new methods have arisen to ensure that Optical Character Recognition (OCR) remains coherent and usable in the creation of better text corpora across a variety of domains and historical periods. This short paper investigates one such effort: the ability of large language models (LLMs) to amend poor transcriptions of 19th-century French monographs. We evaluate the state-of-the-art Llama-3.2-Instruct model, known for its advanced instruction-following capabilities and adaptability to domain-specific fine-tuning, on both real and synthetic data derived from the ICDAR 2017 corpus, enhanced using the 'ScrambledText' module to introduce realistic OCR errors. Although advanced in instruction-following, the results reveal challenges in generalization, with fine-tuned models increasing CER errors, highlighting limitations in LLMs for handling character-level OCR corrections
Conversion of non-food biomass to amphiphilic molecules by mechanochemistry
No full textInternational audienc
F.A.U.V.E. PROJECT - NADES as innovative adjuvant solutions for plant treatment
No full textInternational audienceTo limit the impact on the environment and the risk to humans, the development of new agricultural practices is emerging worldwide, aiming to considerably reduce the amount of sprayed plant treatment, for example, though the use of drones and robots [1]. However, a key challenge remains: how to maintain the stability of the agricultural mixture and the quality of spraying, while significantly reducing the volume of crop treatment solutions and significantly increasing the concentration of active substances? In this context, the FAUVE project (Formulation of innovative Adjuvant solutions for Ultra-low Volume applied by new Equipment) aims to support this agricultural transition in France, by developing innovative adjuvant solutions to maintain the efficacy of concentrated and sprayable plant treatment products for organic agriculture. The dedicated scientific literature increasingly refers to the application potential of NADES (Natural Deep Eutectic Solvents) for their solubilizing power [2]. This study investigates the use of some NADES, such as the combination of choline chloride with glucose or glycerol, to significantly increase the concentration of active substances, such as potassium phosphonate and copper sulfate, in spray mixtures. In addition, to ensure that the formulations remain suitable for spraying, the study examines the effect of adding water to reduce viscosity [3]
Une formalisation en Rocq des éléments finis de Lagrange simpliciaux
No full textThe finite element method is a popular method to numerically solve partial differential equations. In the long-term goal of proving its correctness, we focus here on the formal definition of what is a finite element: a record in the Rocq proof assistant with both values and proofs of validity, including the main one called unisolvence. We then instantiate this record with the most popular and useful, the simplicial Lagrange finite elements for evenly distributed nodes, for any dimension and any polynomial degree. These proofs require many results (definitions, lemmas, canonical structures) about finite families, affine spaces, multidimensional polynomials, in the context of finite or infinite-dimensional spaces.La méthode des éléments finis est un outil populaire pour la résolution numérique d'équations aux dérivées partielles. En vue de l'objectif à long terme de la preuve de sa correction, nous nous focalisons ici sur la définition formelle de ce qu'est un élément fini : un enregistrement dans l'assistant de preuve Rocq avec à la fois des valeurs et des preuves de validité, dont la principale appelée unisolvance. Puis nous instancions cet enregistrement avec les plus populaires et utiles, les éléments finis simpliciaux de Lagrange pour des nœuds uniformément répartis, pour toutes les dimensions et tous les degrés de polynômes. Ces preuves exigent de nombreux résultats (définitions, lemmes, structures canoniques) sur les familles finies, les espaces affines, les polynômes multivariés, dans le cadre d'espaces de dimension finie ou infinie
Adsorption of an active molecule on the surfaces of kaolinite for controlled release applications
No full textInternational audienc
Voltage-Controlled SoC Estimation in Lithium-Ion Batteries: A Comparative Analysis of Equivalent Circuit Models
No full textInternational audienceThis paper introduces voltage-controlled methods to regulate battery voltage and extract state of charge (SoC) in lithium-ion batteries, comparing three equivalent circuit models (ECMs): a simple resistance model, an RC model, and an RC-Diffusion model. The models are tested under various conditions and compared against lab data, evaluating their accuracy
Modular multilevel converter based on unipolar-current full-bridge submodules for medium voltage variable speed drive applications
No full textInternational audienceWith the continuous development of Modular Multilevel Converters (MMC), the introduction of new Sub-Modules (SMs) and their operational characteristics has become favourable for optimizing the performance of many applications such as Variable Speed Drives (VSD), where the conventional ac/dc/ac MMC converters based on Half Bridge (HB) face challenges in lowfrequency operation. This paper focuses on the main features of the Unipolar Current (UC) Full-Bridge (FB) SM (UC FB-SM) and benefits of its integration into the MMCs for VSD applications. Basic operation methods of the HB and UC FB based MMCs are analysed and compared, and the duality between their operational modes is highlighted. Main advantages of utilization of the UC FB MMCs in VSD applications are identified and illustrated in the simulation results presented
Development of an Integrated Transmission-Distribution Simulation Model for Coordinated TSO-DSO Interface Management
No full textInternational audienceThis paper presents a comprehensive methodology for integrated TSO/DSO modeling, aimed at enhancing coordinated management of transmission and distribution networks. The proposed approach allows to perform a unified numerical load-flow model that effectively integrates both transmission and distribution electrical systems, accounting for technical constraints of each grid operator. The validation of the detailed joint TSO/DSO model is carried out using real-world scenarios provided by grid operators, ensuring its practical applicability