Portail HAL Ensta
Not a member yet
11080 research outputs found
Sort by
Shipborne GNSS atmospheric water vapor retrieval for SWOT radiometer validation in the framework of the Wem-SWOT and BioSWOT campaigns
No full textInternational audienceThe Surface Water and Ocean Topography (SWOT) mission allows sub-mesoscale ocean processes to be investigated with great precision. The SWOT KaRin capability to access oceanic scales of about 20 km involves new error sources that were orders of magnitude smaller for classical altimetry. In particular, small-scale atmospheric effects may affect the sea surface height (SSH) product. Water vapor is measured through two radiometers onboard the satellite, measuring at the center of each of the two KaRIn interferometers on each side of the Swot nadir. Hay et al. (2025) highlighted some irregularities in the KaRIn dataset due to the spatial variations of the wet troposphere correction not sensed by the wide footprint and single measurement correction applied to both KaRIn swaths. This research is conducted within the framework of the oceanographic campaigns WemSWOT, led by the Shom, and BioSWOT-2023, led by Ifremer, in March, April and May 2023, involving the research vessel L'Atalante carrying a geodetic GNSS antenna. L'Atalante was sailing along the KaRIn footprint every day during 20 days of the CalVal period of the SWOT mission. We propose to use the GNSS retrieval of water vapor content with state of the art method to qualify the SWOT radiometer measurement of the water vapor used for the correction of the altimetric measurements of SWOT. Indeed, the water vapor sensing of the atmosphere from shipborne geodetic GNSS antennas has been assessed in the literature with an accuracy reaching 2 kg.m-². Assessing the radiometer accuracy in the coastal area, ~50 km from the coast, where the radiometer is less performant, is especially relevant. The coastal effect will be thoroughly documented.To achieve this, we compare the SWOT radiometer dataset to the shipborne GNSS-retrieved water vapor and ERA5 reanalysis from ECMWF. Taking into account the time and location difference between SWOT and in situ GNSS measurements, we evaluate the capabilities of the SWOT radiometer, which could introduce small-scale atmospheric variations into the SSH product
A comparative review of decision-making approaches for realistic event-driven environments
No full textInternational audienceDecision-making in complex, multi-process, event-driven environments, where information arrives as asynchronous event streams, poses distinct challenges compared to traditional scenarios. Factors such as concurrency, asynchronicity, partial observability, and complex interdependencies introduce significant modeling and computational difficulties and require thorough analysis and characterization. This paper offers a comparative review of existing decision-making approaches designed to address these challenges. Leveraging Endsley's model of Situation Awareness, our analysis is structured around the cognitive processes of perception, comprehension, and projection of information. We examine various methodologies, including Temporal Planning and Modeling, Discrete Event Dynamic Systems, Event Processing, and Probabilistic Graphical Models, and identify their strengths, limitations, and applicability to complex, dynamic settings. Our findings underscore the necessity of a comprehensive framework that integrates the decision-action-perception loop by combining the perception of multimodal, semantically enriched data, the comprehension through online learning of stochastic event models emitted by multiple long-range processes, and the projection via decision-making that balances task performance with continuous model refinement. This highlights the value of hybrid solutions that combine the complementary strengths of different approaches to address the multifaceted challenges inherent across all levels of situational awareness and decision-making
A Decade of Research on Hybrid Cloud Storage Systems: A Retrospective
No full textInternational audienc
Optimal Control Problem Under Signal Temporal Logic Constraints: A Robust Reformulation using Augmented Dynamics
No full textThis work presents a novel approach for solving optimal control problems under Signal Temporal Logic (STL) constraints. The proposed method reformulates the original problem as a classical continuous-time optimal control problem by augmenting the system dynamics with auxiliary variables that encode STL satisfaction through their evolution and boundary conditions. Introducing a robustness parameter, we also establish the convergence of the reformulated problem to the original one as this parameter tends to zero. Numerical simulations are realized to demonstrate the feasibility of our method, highlighting its potential for practical applications
How to introduce an initial crack in phase field simulations to accurately predict the linear elastic fracture propagation threshold?
No full textInternational audienceVariational phase field fracture models are now widely used to simulate crack propagation in structures. A critical aspect of these simulations is the correct determination of the propagation threshold of pre-existing cracks, as it highly relies on how the initial cracks are implemented. While prior studies briefly discuss initial crack implementation techniques, we present here a systematic investigation. Various techniques to introduce initial cracks in phase field fracture simulations are tested, from the crack explicit meshing to the replacement by a fully damaged phase field, including different variants for the boundary conditions. Our focus here is on phase field models aiming to approximate, in the -convergence limit, Griffith quasi-static propagation in the framework of Linear Elastic Fracture Mechanics. Therefore, a sharp crack model from classic linear elastic fracture mechanics based on Griffith criterion is the reference in this work. To assess the different techniques to introduce initial cracks, we rely on path-following methods to compute the sharp crack and the phase field smeared crack solutions. The underlying idea is that path-following ensures staying at equilibrium at each instant so that any difference between phase field and sharp crack models can be attributed to numerical artifacts. Thus, by comparing the results from both models, we can provide practical recommendations for reliably incorporating initial cracks in phase field fracture simulations. The comparison shows that an improper initial crack implementation often requires the smeared crack to transition to a one-element-wide phase band to adequately represent a displacement jump along a crack. This transition increases the energy required to propagate the crack, leading to a significant overshoot in the force-displacement response. The take-home message is that to predict the propagation threshold accurately and avoid artificial toughening; the crack must be initialized either setting the phase field to its damage state over a one-element-wide band or meshing the crack explicitly as a one-element-wide slit and imposing the fully cracked state on the crack surface
SCvx-Frank-Wolfe Algorithm
No full textInternational audienceConvex optimization algorithms are increasingly used in embedded trajectory planning for safety-critical systems, requiring verification to meet safety standards. Previous research has shown that formal verification methods can be applied to convex solvers for linear problems. This paper explores a specific implementation of the successive convexification approach for trajectory planning using linear programming. We focus on a new implementation of the successive convexification algorithm based on the Frank-Wolfe method (FW-SCVX), aiming at demonstrating that it maintains algorithm performance and is suitable for embedded systems. We also analyze the convergence of the FW-SCVX algorithm and provide numerical evaluations to support our proposal. This work contributes to the formal verification of advanced trajectory planning algorithms in safety-critical embedded systems
DisPEED: Distributing Packet flow analyses in a swarm of heterogeneous EmbEddeD platforms
No full textInternational audienceSecurity is a major challenge in swarm of drones. Network intrusion detection systems (IDS) are deployed to analyze and detect suspicious packet flows. Traditionally, they are implemented independently on each drone. However, due to heterogeneity and resource limitations of drones, IDS algorithms can fall short in satisfying Quality of Service (Qo S) metrics, such as latency and accuracy. We argue that a drone can make profit from the swarm by delegating part of the analysis of their packet flows to neighbor drones that have more processing power to enforce security. In this paper, we propose two solving methods to distribute the packet flows to analyze among drones in a way to ensure that it is processed with a minimum communication overhead to limit the attack surface, while ensuring Qo S metrics imposed by the drone mission. First, we propose a formulation of the distribution problem using both an Integer Linear Programming (ILP) and a Maximum-Flow Minimum-Cost (MFMC). Furthermore, we propose two specific solving methods for the distribution problem: (1) a Greedy Heuristic (GH), a non-exact solving method, but with small time overhead, and (2) an Adapted Edmonds-Karp (AEK) algorithm, an exact method, but with a higher time overhead. GH proved to be a very fast solution (up to more than 2000x faster than ILP with Branch and Bound), while AEK solution proved to find the exact solution even when the problem is very difficult
Deep-learning-based detection of underwater fluids in multiple multibeam echosounder data
No full textInternational audienceetecting and locating emitted fluids in the water column is necessary for studying margins, identifying natural resources, and preventing geohazards. Fluids can be detected in the water column using multibeam echosounder data. However, manually analyzing the huge volume of this data for geoscientists is a very time-consuming task. Our study investigated the use of a YOLO-based deep learning supervised approach to automate the detection of fluids emitted from cold seeps (gaseous methane) and volcanic sites (liquid carbon dioxide). Several thousand annotated echograms collected from three different seas and oceans during distinct surveys were used to train and test the deep learning model. The results demonstrate first that this method surpasses current machine learning techniques, such as Haar-Local Binary Pattern Cascade. Additionally, we thoroughly analyzed the composition of the training dataset and evaluated the detection performance based on various training configurations. The tests were conducted on a dataset comprising hundreds of thousands of echograms i) acquired with three different multibeam echosounders (Kongsberg EM302 and EM122 and Reson Seabat 7150) and ii) characterized by variable water column noise conditions related to sounder artefacts and the presence of biomass (fishes, dolphins). Incorporating untargeted echoes (acoustic artefacts) in the training set (through hard negative mining) along with adding images without fluid-related echoes are the most efficient way to improve the performance of the model and reduce the false positives. Our fluid detector opens the door for near-real time acquisition and post-acquisition detection with efficiency, reliability and rapidity
Unraveling Major Questions in Micronekton Ecology and Their Role in the Biological Carbon Pump Through Integrative Approaches and Autonomous Monitoring
No full textInternational audienceMicronekton consist of crustaceans, cephalopods, gelatinous organisms, and fishes that are 2–20 cm in size (Figure 1). These organisms have unique functional traits that impact their vertical migration patterns and ecosystem processes (Aparecido et al., 2023). Our understanding of their potential carbon transport and sequestration from the epipelagic (upper 200 m) to mesopelagic zones (200–1,000 m) or deeper (e.g., Boyd et al., 2019; Le Moigne, 2019; Cavan et al., 2019) is limited by the tools traditionally used to assess their biomass, diversity, and varied migration patterns (e.g., Annasawmy et al., 2019, 2024; Barbin et al., 2024; Eduardo et al., 2024). These knowledge gaps are notable considering that micronekton are ubiquitous throughout the world ocean
Verification and evaluation of computer and communication systems
No full textInternational audienceThis volume contains the papers presented at the 17th International Conference on Verification and Evaluation of Computer and Communication Systems (VECoS 2024), held during October 16–18, 2024 in Djerba, Tunisia.The event of this year continues the tradition of previous editions held 2007 in Algiers, 2008 in Leeds, 2009 in Rabat, 2010 in Paris, 2011 in Tunis, 2012 in Paris, 2013 in Florence, 2014 in Bejaïa, 2015 in Bucharest, 2016 in Tunis, 2017 in Montreal, 2018 in Grenoble, 2019 in Porto, 2020 in Xi’an (virtual), 2021 in Beijing (virtual), and 2023 in Marrakech.As in previous editions, VECoS provided a forum for researchers and practitioners in the areas of verification, control, performance, and dependability evaluation in order to discuss the state of the art and challenges in modern computer and communication systems in which functional and extra-functional properties are strongly interrelated. The main motivation is to encourage the cross-fertilization between various formal verification and evaluation approaches, methods, and techniques, and especially those developed for concurrent and distributed hardware/software systems.The Program Committee of VECoS 2024 was composed of 77 researchers from 21 countries. We received 42 full submissions from 7 countries. After a thorough and lively discussion phase, the committee decided to accept 16 regular papers. The topics presented covered a range of subjects, including approaches to improving the scalability and efficiency of formal verification and its applications to blockchain, smart contracts, and neural networks. The conference also included two invited talks, one on fault diagnosis of discrete-event systems using Petri nets and the other on descriptive and prescriptive system models formalised in Event-B.We are grateful to the Program and Organizing Committee members, to the reviewers for their cooperation, and to Springer for their professional support during the production phase of the proceedings. We are also thankful to all authors of submitted papers, to the invited speakers, and to all participants of the conference. Their interest in this conference and contributions are greatly appreciated