Archive ouverte de l'ENAC
Not a member yet
3458 research outputs found
Sort by
Effects of one session of theta or high alpha neurofeedback on EEG activity and working memory
No full textInternational audienceNeurofeedback techniques provide participants immediate feedback on neuronal signals, enabling them to modulate their brain activity. This technique holds promise in unveiling brain-behavior relationship, and offers opportunities for neuroenhancement. Establishing causal relationships between modulated brain activity and behavioral improvements requires rigorous experimental designs, including appropriate control groups and large samples. Our primary objective was to examine whether a single neurofeedback session, designed to enhance working memory through the modulation of theta or high-alpha frequencies, elicits specific changes in electrophysiological and cognitive outcomes. Additionally, we explored predictors of successful neuromodulation. One hundred and one healthy adults were assigned to groups trained to increase frontal theta, parietal high alpha, or random frequencies (active control group). We measured resting-state EEG, working memory performance, and self-reported psychological states before and after one neurofeedback session. Although our analyses revealed improvements in electrophysiological and behavioral outcomes, these gains were not specific to the experimental groups.An increase in the frequency targeted by the training has been observed for the theta and high alpha groups, but training aimed at increasing randomly selected frequencies appears to induce more generalized neuromodulation compared to targeting a specific frequency. Among all the predictors of neuromodulation examined, resting theta and high alpha amplitudes predicted specifically the increase of those frequencies during the training. These results highlight the challenge of integrating a control group based on enhancing randomly selected frequency bands and suggest potential avenues for optimizing interventions (e.g., by including a control group trained in both up-and down-regulation).</p
Hovering stabilization of the DarkO tail-sitter drone with constant wind
No full textWe present two mathematical models of the DarkO tail-sitter convertible UAV developed and 3D printed at the École Nationale de l'Aviation Civile (ENAC), in Toulouse (France). During a hover flight, the UAV is vertical, which offers a large wing area facing the wind. Thus, aerodynamic disturbances have a strong influence on the nonlinear UAV dynamics. Our models capture this behavior and allow us to characterize relevant equilibria, in the presence of a constant wind, and the corresponding wind-dependent linearized dynamics. Using a parametric family of models, we design an optimality-based robust static output feedback controller that uses two degrees of freedom on the orientation to stabilize the UAV in a hovering condition in spite of an unknown constant wind. This control law has been implemented in the Paparazzi autopilot software to obtain experimental results that validate our theory.</div
Enhancing airline connectivity: An optimisation approach for flight scheduling in multi-hub networks with bank structures
No full textInternational audienceWhile one salient characteristic of hub airports lies in connecting passengers, the full-service airlines in North America concentrate their networks spatially over a number of hubs. Having witnessed the emerging multi-hub network, this paper investigated the flight scheduling problem under a multi-hub configuration, taking a well-defined bank structure and airport operational restrictions into account. An integrated non-convex Mixed-Integer Nonlinear Programming (MINLP) approach was proposed to enhance airline connectivity, considering different combinations of traffic flow direction and connecting times. To verify the scalabilityand effectiveness of the proposed model, a comprehensive case study has been undertaken with real-world scheduling data from Air China, which was solved by a novel problem-specific Selective Simulated Annealing (SSA) algorithm. Substantial improvements were achieved without sacrificing the scheduling efficiency. Precisely, the program adjusted the flights during a typical operational day in a timely manner. The post-optimisation outcomes have witnessed its effectiveness with a 17.97%, 17.06%, 22.41% and 53.86% increase in airline connectivity at its four major hub airports (Chengdu Shangliu, Beijing Capital, Shanghai Pudong and Hongqiao) in China, respectively. A clear pattern of the bank structure also confirms its positive impact on airline connectivity under the multi-hub network configuration. Lastly, a comparative analysis for the distribution of all feasible connections further highlights the critical challengeconcerning the role of the hubs in a multi-hub network. More specifically, Air China’s multi-hub network systematically performs better on Domestic-International routes, due to flight schedule, frequencies, geographical placements and detours. Among the four hub airports, Beijing Capital International Airport stands out as a dominant one, which implies its potential to serve as a robust international hub airport
Distributional loss for convolutional neural network regression and application to parameter estimation in satellite navigation signals
No full textInternational audienceConvolutional Neural Network (CNN) have been widely used in image classi-cation. Over the years, they have also beneted from various enhancements andthey are now considered state-of-the-art techniques for image-like data. However, when they are used for regression to estimate some function value fromimages, few recommendations are available to construct robust CNN regressormodels. In this study, a robustness enforcing mechanism is proposed for CNNregression models. It combines convolutional neural layers to extract high levelfeatures representations from images with a soft labelling technique that helpsgeneralization performance. More specically, as the deep regression task ischallenging, the idea is to account for some uncertainty in the targets that areseen as distributions around their mean. Building from earlier work (Imani &White, 2018), a specic histogram loss function based on the Kullback-Leibler(KL) divergence is applied during training. The prior distributions are selectedaccording to the physical characteristics of the parameters to estimate. To assess and illustrate the technique, the model is applied to Global NavigationSatellite System (GNSS) multipath estimation where multipath signal parameters have to be estimated from correlator output images from the I and Qchannels. The multipath signal delay, magnitude, Doppler shift frequency andphase parameters are estimated from synthetically generated datasets of satellite signals. Experiments are conducted under various receiving conditions andvarious input images resolutions to test the estimation performances qualityand robustness. The results show that the proposed soft labelling CNN technique using distributional loss outperforms classical CNN regression under allconditions. Furthermore, the extra learning performance achieved by the mode
How a pilot’s brain copes with stress and mental load? Insights from the executive control network
No full textInternational audienceIn aviation, mental workload and stress are two major factors that can considerably impact a pilot’s flight performance and decisions. Their consequences can be even more dramatic in single-pilot aircraft or with the forthcoming single-pilot operations where the pilot will fly alone and will not be able to be assisted in case of difficulty. An accurate and automatic monitoring of the pilot’s mental state could help to prevent the potentially dangerous effects of an excess mental workload and stress. For example, some tasks could be allocated to automation or to a ground-based flight crew if a mental overload or significant stress is detected. In the current study, the brain activity of 20 private pilots was recorded with a fNIRS device during two realistic flight simulator scenarios. The mental workload was manipulated with the added difficulty of a secondary task and stress was induced by a social stressor. Our results confirmed the sensitivity of the fNIRS readings to variations in the mental workload, with increased HbO2 concentration in regions of the executive control network (ECN), in particular in the dorsolateral prefrontal cortex and in lateral parietal regions, when the difficulty of the secondary task was high. The social stressor also triggered an HbO2 increase in the ECN, especially when it was combined with high mental workload. This latter result suggests that mental workload and stress together can have cumulative effects, and coping with both factors is possible at the expense of an extra recruitment of the ECN. Finally, results also revealed a time-on-task effect, with a progressive reduction of the HbO2 signal in the ECN during the flight scenario, suggesting that these regions are sensitive to short term habituation to the tasks. Overall, fNIRS efficiently indexed mental load, stress, and practice effects
Comet Interceptor: An ESA Mission to a Yet Unidentified Target
No full textInternational audienceAs of today, 8 different comets have been visited by 10 spacecraft. These missions have provided invaluable knowledge about comet compositions, shapes, and structure. However, all these visited comets had previously completed multiple orbits around the Sun. The predictability of their return is clearly a key aspect for their accessibility, including the planning, development and launch of missions capable of intercepting them. However, the multiple passages through the inner solar system substantially alter the morphological and chemical characteristics of comets, hindering the understanding of their formation. ESA F-Class mission Comet Interceptor (Comet-I) aims to visit a comet that has yet to enter the inner solar system and, as such, is also yet to be discovered. In 2029, Comet-I will share a ride with ESA’s M4 ARIEL to a quasi-Halo orbit around the Sun-Earth Lagrange 2 point (SEL-2). The coupling of the orbital energy at SEL-2 together with its dynamical instability allows for Comet-I to depart the Earth gravity well with an excess escape velocity in the range from 0.8 to 1 km/s. Added to this, Comet-I propulsion system can perform a total of 600 m/s of Δv change. Given this orbital manoeuvrability, the paper presents an accessibility analysis to all historic long period comets discovered since 1st January 2000. In particular, a subset of 89 objects that crossed the ecliptic plane at heliocentric distances between 1.5 and 0.7 AU is of particular interest as these represent comets with very similar characteristics to those targeted by Comet-I. Out of these, 37 comets would have been reachable by Comet-I if they would have been discovered with warning times ranging 1 to 4 years. The Vera C. Rubin Observatory’s Legacy Survey of Space and Time (LSST) is expected to provide comparable warning times, and so, enable the capability to reach undiscovered objects with modest Δv capabilities. This paper describes the current work at the Comet-I Target Identification Working Group to assess the likelihood of different discovery scenarios, in an effort to inform the decision-making process of whether to intercept the first available newly-discovered long period comet or wait for a potentially more scientifically compelling object.<br /
Surrogate-Based Modeling and Optimization of Expensive System Architectures Problems using ONERA and DLR Softwares: An Integrated Framework
No full textInternational audienceFor developing innovative systems architectures, modeling and optimization techniques havebeen central to frame the architecting process and define the optimization and modeling prob-lems. In this context, for system-of-systems the use of efficient dedicated approaches (oftenphysics-based simulations) is highly recommended to reduce the computational complexity of thetargeted applications. However, exploring novel architectures using such dedicated approachesmight pose challenges for optimization algorithms, including increased evaluation costs andpotential failures. To address these challenges, surrogate-based optimization algorithms, such asBayesian optimization utilizing Gaussian process models have emerged. The proposed solutionrelies mainly on leveraging the capabilities of ONERA and DLR software as well as otheropen-source solutions. In fact, for this purpose we leverage the use of the ONERA softwareSEGOMOE (Super-Efficient Global Optimization with Mixture of Experts) based on surro-gate models from SMT (Surrogate Modeling Toolbox) to handle hierarchical variables. Wealso used the DLR software SBArchOpt to interface optimizers with architecture models likeOpenTurbofanArchitecting. The hierarchical variables support in SMT enables an effectiverepresentation of design decisions, enhancing optimization outcome. In particular, this paperdemonstrates SEGOMOE capabilities to help solving realistic architecture aircraft engine testproblems that are defined within the DLR OpenTurbofanArchitecting software. Throughempirical evaluations and case studies, we demonstrate the effectiveness of our integratedapproach in optimizing jet engine architecture design under real-world constraints, includ-ing hidden constraints. This work contributes to advancing the field of system architectureoptimization and offers valuable insights for future research in surrogate-based optimizationtechniques
Multipath Separation in Split-Step Simulation of Electromagnetic Waves in the Low Troposphere
No full textInternational audienceThe impact of multipath effects on aviation navigation and communication systems is significant. This study aims to understand the multipath effects and mechanisms of electromagnetic waves propagating within complex environments. It aims to accurately split electromagnetic waves, computed via split-step simulation methods, into distinct groups with various group delays. The paper details a modeling technique that addresses multipath effects and a method for separating group delays. A numerical experiment showcases electromagnetic wave propagation in a standard atmosphere over an impedance ground, demonstrating successful separation into multiple wave components with diverse path delays, validating the presented methodologies' efficiency
Vers un outil adaptatif d'aide à la résolution de conflits pour le contrôle aérien.
No full textThe aim of this thesis is to develop a decision support tool to assist air traffic controllersin their daily traffic management in the context of less structured air routes. The use ofsuch a tool would allow one to increase the traffic density by reducing controllers conflictresolution task, and to reduce the delays caused by aircraft avoidance by incorporatingflexible criteria into the choice of manoeuvres.In the past, evolutionary algorithms have demonstrated their effectiveness in resolving airtraffic conflicts in a static context, with two main advantages : their ability to providepopulations of solutions rather than a single solution, and to allow flexible trajectorymodels.This thesis addresses two related problems. The first part of the thesis proposes the use ofan evolutionary algorithm in a dynamic context to resolve airspace conflicts in a controlsector. In a first step, operators of the evolutionary algorithm are modified. Solutionsobtained are compared to solutions provided by an exact constraint programming methodto verify that their quality is equivalent. In order to propose a continuous solution, anaive approach and an approach using a memory preserving the previously computedsolutions of the algorithm are presented. Results obtained from the simulations highlightthe advantages of using explicit memory and new operators to speed up successive solutionswithout affecting the quality and diversity of the answers provided. Furthermore, anddespite the reuse of the same population of individuals, this approach proves to be robustwhen unexpected new constraints happen in the sector, which could simulate the decisionsof air traffic controllers.The second part consists of learning and interpreting the uncertainty parameters of airtraffic controllers in their aircraft conflict resolution task. Learning these uncertainties canhelp to calibrate the decision support tool as closely as possible to the operating modesof human controllers. This step is essential to make the tool acceptable and useful. Thelearning method introduced using an evolutionary algorithm was first validated on nominaldata and then tested on experimental data. The integration of the parameters found intoa conflict resolution algorithm led to solutions that are more similar to human solutions.Finally, we apply the method to real traffic data recorded at the Bordeaux Control Centre.This last step provided orders of magnitude for the uncertainty parameters.La motivation de cette thèse est de développer un outil d’aide à la décision pour assisterles contrôleurs aériens au quotidien dans leur gestion du trafic et dans un contexte deroutes aériennes moins structurées. L’utilisation d’un tel outil permettrait d’augmenter ladensité acceptable de trafic en allégeant la charge de résolution de conflits des contrôleurset de réduire les retards engendrés par les évitements entre avions en intégrant des critèresflexibles dans le choix des manoeuvres.Les algorithmes évolutionnaires ont démontré par le passé leur efficacité dans la résolutionde conflits aériens dans un contexte statique, avec deux avantages principaux : leur capacitéà fournir des populations de solutions et non une solution unique, ainsi que la souplessedu modèle de trajectoires qu’ils permettent d’envisager.Le travail de cette thèse se divise en deux axes. Le premier propose d’utiliser un algorithmeévolutionnaire dans un contexte dynamique pour résoudre des conflits aériens dans unsecteur de contrôle. Une première étape modifie les opérateurs de l’algorithme évolutionnaireet vérifie que l’algorithme obtient des solutions de qualité équivalente à des résolutionsexactes fournies par un algorithme de programmation par contraintes. Dans le but deproposer une résolution continue, une approche naïve et une approche avec mémoireconservant les solutions précédemment calculées sont présentées. Les résultats obtenus àl’issue de simulations soulèvent les bénéfices de la mise en place d’une mémoire explicite etdes nouveaux opérateurs pour accélérer les résolutions successives sans impacter la qualitéet la diversité des réponses apportées. De plus, et malgré la réutilisation d’une mêmepopulation d’individus, cette approche se montre robuste face à des contraintes ponctuellesintroduites au sein du secteur qui pourraient simuler des décisions des contrôleurs aériens.Le second axe consiste à apprendre et à interpréter les paramètres d’incertitude descontrôleurs aériens lors de leur tâche de résolution de conflits entre aéronefs. L’apprentissagede ces incertitudes a pour objectif de calibrer au mieux l’outil d’aide à la décision encohérence avec les modes opératoires des contrôleurs humains, ce qui est essentiel pour lerendre acceptable et utile lors de son utilisation. La méthode d’apprentissage introduite àl’aide d’un algorithme évolutionnaire a été premièrement validée sur des données nominales.Des tests ont ensuite été réalisés sur des données expérimentales, et l’intégration desparamètres d’incertitude ainsi appris à un algorithme de résolution de conflits a permisd’aboutir à des solutions proposées davantage similaires aux résolutions humaines. Enfin,la méthode a été appliquée sur des données de trafic réel enregistrées au centre de contrôlebordelais. Cette dernière étape a fourni des ordres de grandeur des paramètres d’incertitude