HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
11652 research outputs found
Sort by
Safe and Adaptive Roundabout Insertion for Autonomous Vehicle based on Limit-cycle and Predicted Inter-Distance Profiles
No full textInternational audienceRoundabouts are a prevalent form of road infrastructure that effectively control traffic flow and significantly decrease the occurrence of accidents in contrast to traditional intersections. This paper, based on the Multi-Risk Assessment and Management Control Strategy (MRAM-CS) [1] aims to enhance this architecture by considering the obstacle behavior. The MRAM-CS allows autonomous vehicles (called Ego-Vehicles (EVs) in what follows) to determine whether to accelerate or decelerate at the arrival of the roundabout and to enter by applying an appropriate speed profile, determined online, which allows to respect appropriate distances with the vehicles circulating in the roundabout. This is done by using the Predictive Inter-Distance Profile metric (PIDP) and the dynamic progress of the minimum value of PIDP (mPIDP). The proposed control is based on Fuzzy-PID controller, allowing to update the PID gains according to Fuzzy Inference System (FIS) and the behavior feature (calm, aggressive or dangerous) of the other vehicles. Several simulations are performed to demonstrate the reliability and the safety of the proposed approach
An approach to optimize the railway resource planning process in a microscopic scale
No full textInternational audienc
XGBRegressor and tree-structured Parzen estimator based approach for predicting proton exchange membrane fuel cell performance of electro-hydrogen
No full textInternational audienceThe proton exchange membrane fuel cell (PEMFC) is a critical and essential component of a zeroemissionelectro-hydrogen generator. The accurate prediction of its performance is important for optimal operationmanagement and preventive maintenance of the system. However, the prediction is not easy because thePEMFCs have complex electrochemical reactions with multiple nonlinear relations between operating variablesas inputs and voltage as output. In this paper, we propose an efficient prediction approach based onXGBRegressor and Tree-structured Parzen Estimator. In addition, to better select relevant features, KernelPrincipal Component Analysis and Mutual Information are jointly used. The proposed approach allows consideringthe dynamic operating conditions of the fuel cells. To test and validate the robustness of the proposedapproach, a data-set of ten PEMFCs was used. Furthermore, a comparison study with traditional machinelearning models, such as Artificial Neural Networks and Support Vector Machine Regressors is investigated. Itwas shown that the proposed approach provides better results
Nonconventional Angle-Of-Attack Control Strategy for Reducing the Airspeed During the Fixed-Wing Drone Landing
No full textInternational audienceIn this work, a control strategy for landing a fixed-wing drone with classical configuration and with minimum airspeed on a touchdown point is presented. In this crucial and critic landing phase of this kind of aircrafts the challenge is to absorb the drone’s airspeed without loss its controllability. Our strategy proposes a scientific solution for a safe landing by controlling the angle of attack of the drone assuring its stability during all this stage. In our analysis, a flight scheme composed by a cruise flight and a landing scheme is considered. The control strategy obtained from the Lyapunov theory proposes a critic descending angle to obtain a maximum airspeed reduction. In addition, an observer is proposed for estimating external aerodynamics parameters and compensate them in closed-loop system. Numerical validation corroborates the well performance of the proposed control algorithms
Fucoidan-coated coils improve healing in a rabbit elastase aneurysm model
No full textInternational audienceBackground Recanalization of coiled aneurysms remains unresolved. To limit aneurysm recanalization after embolization with coils, we propose an innovative approach to optimize aneurysm healing using fucoidan-coated coils. Objective To evaluate the short-term efficacy and long-term safety of the new coil system with conventional angiography, histology, and multiphoton microscopy for follow-up of fibrosis and neointima formation. Methods We conducted a feasibility study on rabbit elastase-induced aneurysms. Embolization was carried out with bare platinum coils, fucoidan-coated coils, or dextran-coated coils. Aneurysms were controlled after 1 month by digital subtraction angiography (DSA). Aneurysm samples were collected and processed for histological analysis. Aneurysm healing and fibrosis were measured by quantifying collagen according to the histological healing score by combining standard light microscopy and multiphoton imaging. We divided 27 rabbits into three groups: bare platinum group, fucoidan group, and dextran group as controls. Results Angiographic grading showed a trend toward less recanalization in the fucoidan group, although there were no significant differences among the three groups (P=0.21). Histological healing was significantly different according to the presence of more collagen in the neck area of aneurysms in the fucoidan group versus the bare platinum group (P=0.011), but not in the dextran group. Histological index was significantly better at the aneurysm neck in the fucoidan group than in the bare platinum group (P=0.004). Collagen organization index was also significantly better in the fucoidan group than in the bare platinum group (P=0.007). Conclusion This proof-of-concept study demonstrated the feasibility and efficacy of treatment with fucoidan-coated coils to improve aneurysm healing. The results in this rabbit in vivo model showed that fucoidan-coated coils have the potential to improve healing following endovascular treatment
Extensions of the empirical interpolation method to vector-valued functions
No full textIn industrial Computer-Assisted Engineering, it is common to deal with vector fields or multiple field variables. In this paper, different vector-valued extensions of the Empirical Interpolation Method (EIM) are considered. EIM has been shown to be a valuable tool for dimensionality reduction, reduced-order modeling for nonlinear problems and/or synthesis of families of solutions for parametric problems. Besides already existing vector-valued extensions, a new vector-valued EIM-the so-called VEIM approach-allowing interpolation on all the vector components is proposed and analyzed in this paper. This involves vector-valued basis functions, same magic points shared by all the components and linear combination matrices rather than scalar coefficients. Coefficient matrices are determined under constraints of point-wise interpolation properties for all the components and exact reconstruction property for the snapshots selected during the greedy iterative process. For numerical experiments, various vector-valued approaches including VEIM are tested and compared on various one, two and three-dimensional problems. All methods return robustness, stability and rather good convergence properties as soon as the Kolmogorov width of the dataset is not too big. Depending of the use case, a suitable and convenient method can be chosen among the different vector-valued EIM candidates
Electrospinning of ultrafine non-hydrolyzed silk sericin/PEO fibers on PLA : a bilayer scaffold fabrication
No full textInternational audienceWe report the feasibility of electrospinning of protein-polymer multilayered scaffolds with selected materials such as non-hydrolyzed silk sericin (SS), poly- ethylene oxide (PEO), and polylactic acid (PLA), with tuned fiber size and properties for each layer. We present a new innovative way for the electrospin- ning (ES) of non-hydrolyzed SS mixed with PEO yielding fibers with an aver- age diameter ranging between 120 and 150 nm. Different SS:PEO ratios have been electrospun to study the effect of the concentration of SS protein on the fibers size and shape, as well and their electrospinnability. Electrospun SS:PEO fibers display weak to no mechanical resistance (non-measurable) and their deposition onto a sturdier scaffold is necessary to allow their use in bio- medical and/or pharmaceutical fields. Therefore, bilayer scaffolds have been fabricated consisting of a PLA support and SS:PEO fibers obtained from the optimized SS:PEO ratio (1.2:4). They are composed of a sturdy hydrophobic layer of PLA fibers and a layer of sticky hydrophilic SS:PEO fibers. The scaf- folds have been characterized extensively by Fourier transforms infra-red (FTIR) spectroscopy, X-ray diffraction (XRD), scanning electron microscopy (SEM), and their resistance to mechanical stress. Finally, hydrophobicity of both layers has been determined by measuring the contact angle of water drop- lets on the scaffolds, further proving the bilayer nature of the scaffolds
Development of preclinical mri elastography at 7T: an original approach for the non-invasive mechanical characterization of mouse organs
No full textInternational audienceMagnetic Resonance Elastography (MRE) is a non-invasive imaging method that aims at mapping the mechanical properties of a tissue in vivo. MRE is based on 3 steps: (1) the mechanical excitation of the organ by external generation of shear waves; (2) the encoding of the sample response by MRI using specific sequences; (3) the reconstruction of the mechanical properties. In a more marginal way, MRE has been developed for preclinical studies, such as investigation of liver, brain or skeletal muscle pathologies in mice. However, few MRE setups are currently available for preclinical purpose. In the framework of a French multicenter consortium, we have recently developed an innovative and original MRE process to investigate the mechanical properties in different organs of mouse. Available at the IRIS platform of the ICube laboratory in Strasbourg, this research tool combines performances for in vivo mapping of mechanical properties
