Search

International audience; This paper focuses on the problem of the hierarchical multi‐label classification of research papers, which is the task of assigning the set of relevant labels for a paper from a hierarchy, using reduced amounts of labelled training data. Specifically, we study leveraging unlabelled data, which are usually plentiful and easy to collect, in addition to the few available labelled ones in a semi‐supervised learning framework for achieving better performance results. Thus, in this paper, we propose a semi‐supervised approach for the hierarchical multi‐label classification task of research papers based on the well‐known Co‐training algorithm, which exploit content and bibliographic coupling information as two distinct papers' views. In our approach, two hierarchical multi‐label classifiers, are learnt on different views of the labelled data, and iteratively select their most confident unlabelled samples, which are further added to the labelled set. The success of our suggested Co‐training‐based approach lies in two main components. The first is the use of two suggested selection criteria (i.e., Maximum Agreement and Labels Cardinality Consistency) that enforce selecting confident unlabelled samples. The second is the appliance of an oversampling method that rebalances the labels distribution of the initial labelled set, which reduces the reinforcement of the label imbalance issue during the Co‐training learning. The proposed approach is evaluated using a collection of scientific papers extracted from the ACM digital library. Performed experiments show the effectiveness of our approach with regards to several baseline methods.

International audience; Interest and need for Wireless Body Area Networks (WBANs) havesignicantly increased recently. WBAN consists of miniaturized sensors de-signed to collect and transmit data through wireless network, enabling medicalspecialists to monitor patients during their normal daily life and providing realtime opinions for medical diagnosis. Many wireless technologies have provedthemselves in WBAN applications, while others are still under investigations.The choice of the technology to adopt may depend on the disease to moni-tor and the performance requirements, i.e. reliability, latency and data rate.In addition, the suitable sensor is essential when seeking to extract the datarelated to a medical measure.This paper aims at surveying the wireless technologies used in WBAN sys-tems. In addition to a detailed survey on the existing technologies, the use ofthe emerging Low Power Wide Area Network (LPWAN) technologies, and thefuture 5G, B5G and 6G is investigated, where the suitability of these technolo-gies to WBAN applications is studied from several perspectives. Furthermore,medical applications of WBAN are discussed by presenting their methodolo-gies, the adopted wireless technologies and the used sensors. Given that eachmedical application requires the appropriate sensor to extract the data, wehighlight a wide range of the sensors used in the market for medical systems.Recent and future challenges in WBAN systems are given related to the powerconsumption, the emergence of the Internet of Things (IoT) technologies inWBAN and others.

Background In the health care environment, teamwork is paramount, especially when referring to patient safety. We are interested in recent and innovative solutions such as escape games, which is a type of adventure game that may be highly useful as an educational tool, potentially combining good communication skills with successful gamification. They involve teams of 5 to 10 individuals who are “locked” in the same room and must collaborate to solve puzzles while under pressure from a timer. Objective The purpose of this paper was to describe the steps involved in creating and implementing an educational escape game. This tool can then be put into service or further developed by trainers who wish to use it for learning interprofessional collaboration. Therefore, we started with an experience of creating an educational escape game for emergency medicine teams. Methods We chose to develop an educational escape game by using 6 successive steps. First, we built a team. Second, we chose the pedagogical objectives. Third, we gamified (switched from objectives to scenario). Next, we found the human and material resources needed. Thereafter, we designed briefing and debriefing. Lastly, we tested the game. Results By following these 6 steps, we created the first ambulant educational escape game that teaches people, or nurses, doctors, and paramedics, working in emergency medicine to work as a team. Conclusions From a pedagogic point of view, this game may be a good tool for helping people in multidisciplinary fields (medical and paramedical teams) to learn how to work collaboratively and to communicate as a group. Above all, it seems to be an innovative tool that complements medical simulation–based learning and thus consolidates traditional education.

Biodiesel is an eco-friendly, renewable, and potential liquid biofuel mitigating greenhouse gas emissions. Biodiesel has been produced initially from vegetable oils, non-edible oils, and waste oils. However, these feedstocks have several disadvantages such as requirement of land and labor and remain expensive. Similarly, in reference to waste oils, the feedstock content is succinct in supply and unable to meet the demand. Recent studies demonstrated utilization of lignocellulosic substrates for biodiesel production using oleaginous microorganisms. These microbes accumulate higher lipid content under stress conditions, whose lipid composition is similar to vegetable oils. In this paper, feedstocks used for biodiesel production such as vegetable oils, non-edible oils, oleaginous microalgae, fungi, yeast, and bacteria have been illustrated. Thereafter, steps enumerated in biodiesel production from lignocellulosic substrates through pretreatment, saccharification and oleaginous microbe-mediated fermentation, lipid extraction, transesterification, and purification of biodiesel are discussed. Besides, the importance of metabolic engineering in ensuring biofuels and biorefinery and a brief note on integration of liquid biofuels have been included that have significant importance in terms of circular economy aspects. Fil: Chintagunta, Anjani Devi. Vignan’s Foundation for Science, Technology and Research. Department of Biotechnology; India Fil: Zuccaro, Gaetano. Institut National de la Recherche Agronomique; Francia. Università degli Studi di Napoli Federico II; Italia Fil: Kumar, Mahesh. Central Agricultural University; India Fil: Kumar, S. P. Jeevan. Indian Institute of Seed Science; India. Directorate of Floricultural Research; India Fil: Garlapati, Vijay Kumar. Jaypee University of Information Technology; India Fil: Postemsky, Pablo Daniel. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Bahía Blanca. Centro de Recursos Naturales Renovables de la Zona Semiárida. Universidad Nacional del Sur. Centro de Recursos Naturales Renovables de la Zona Semiárida; Argentina Fil: Kumar, N. S. Sampath. Vignan’s Foundation for Science, Technology and Research. Department of Biotechnology; India Fil: Chandel, Anuj K.. Universidade de Sao Paulo; Brasil Fil: Simal Gandara, Jesus. Universidad de Vigo; España

Learning how to model a problem described in natural language as a linear program requires students to practice using various and numerous exercises. Moreover, immediate feedback on the validity of their solutions helps a better and faster understanding. In this paper, we present an idea on how students and teachers can automatically evaluate linear programming models. We also describe how this idea was implemented on the learning platform caseine.org and is now used by hundreds of students from various universities.

International audience; Automation throughout history has caused profound changes in employment dynamics. With the advent of the fourth industrial revolution, a new threat may affect employability, asrobots and AI-based processes can now assume tasks considered exclusive to humans. This position paperaims to motivate the study of the effects of AI and automation on employability, extending it into a collaborative network perspective. The problem is firstly observed from a historical perspective. The collaboration aspects are considered through the analysis of two casestudies. Results suggestthata latent element of collaborative networks, complexity, may have effects in terms of employability.

Background In the recent decades, the number of apps promoting health behaviors and health-related strategies and interventions has increased alongside the number of smartphone users. Nevertheless, the validity process for measuring and reporting app quality remains unsatisfactory for health professionals and end users and represents a public health concern. The Mobile Application Rating Scale (MARS) is a tool validated and widely used in the scientific literature to evaluate and compare mHealth app functionalities. However, MARS is not adapted to the French culture nor to the language. Objective This study aims to translate, adapt, and validate the equivalent French version of MARS (ie, MARS-F). Methods The original MARS was first translated to French by two independent bilingual scientists, and their common version was blind back-translated twice by two native English speakers, culminating in a final well-established MARS-F. Its comprehensibility was then evaluated by 6 individuals (3 researchers and 3 nonacademics), and the final MARS-F version was created. Two bilingual raters independently completed the evaluation of 63 apps using MARS and MARS-F. Interrater reliability was assessed using intraclass correlation coefficients. In addition, internal consistency and validity of both scales were assessed. Mokken scale analysis was used to investigate the scalability of both MARS and MARS-F. Results MARS-F had a good alignment with the original MARS, with properties comparable between the two scales. The correlation coefficients (r) between the corresponding dimensions of MARS and MARS-F ranged from 0.97 to 0.99. The internal consistencies of the MARS-F dimensions engagement (ω=0.79), functionality (ω=0.79), esthetics (ω=0.78), and information quality (ω=0.61) were acceptable and that for the overall MARS score (ω=0.86) was good. Mokken scale analysis revealed a strong scalability for MARS (Loevinger H=0.37) and a good scalability for MARS-F (H=0.35). Conclusions MARS-F is a valid tool, and it would serve as a crucial aid for researchers, health care professionals, public health authorities, and interested third parties, to assess the quality of mHealth apps in French-speaking countries.

International audience; In recent decades and to deal with the scarcity of fossil fuels, many studies have been developed in order to set up a sustainable biofuel production sector. This new sector must be efficient (high productivity), economically profitable (low production costs and therefore acceptable fuel prices), and ethical (low carbon balance, no competition with food resources). The production of bioethanol is based on the fermentation of reserve sugars, accumulated in the form of starch in microalgae and glycogen in cyanobacteria. The advantage of this bioenergy production route lies in the fact that the post-crop fermentation process is at the industrial stage since it has already been tested for many years for the production of bioethanol from agricultural resources. One of the most cultivated cyanobacteria is Arthrospira (“Spirulina”) and its production is also already at industrial scale. Depending on the cultivation conditions, this cyanobacteria is able to accumulate up to 65% DW (dry weight) of glycogen, making it a feasible feedstock for bioethanol production. The aim of this review is to provide a clear overview of these operating conditions for glycogen accumulation.

International audience; The Internet of Things (IoT) is a broad concept comprising a wide ecosystem of interconnected services and devices connected to the Internet. The IoT concept holds fabulous promises, but security aspects tend to be significant barriers for the adoption of large-scale IoT deployments. This paper proposes an approach to assist companies or organisations in the security audit of IoT systems. This approach called Model Learning and Checking Approach (MLCA) combines model learning for automatically extracting models from event logs, and model checking for verifying whether security properties, given under the form of generic LTL formulas hold on models. The originality of MLCA lies in the fact that auditors do not have to craft models or to be expert LTL users. The LTL formula instantiation, which makes security properties concrete, is indeed semi-automatically performed by means of an expert system composed of inference rules. The latter encode some expert knowledge, which can be applied again to the same kind of systems with less efforts. We evaluated MLCA on 5 IoT systems with security measures provided by the European ENISA institute. We show that MLCA is very effective in detecting security issues and provides results within reasonable time.

International audience; The Digital Twin has the ability to store historical and current data about the physical object. In IoT, the concept of general-purpose sensing is aiming at determining a set of basic sensing capabilities from which deriving, by means of Artificial Intelligence algorithms and techniques (termed Synthetic Sensing), relevant information for describing an environment. This paper explores the possible relationships and the feasibility of integration of Synthetic Sensing within a Digital Twin framework. The relevant concepts and associated technologies, the challenges, and some of the enabled scenarios that this integration can bring are presented. This approach is in its infancy and there is a strong need to validate and assess its viability, feasibility, and benefits. The paper identifies some challenges and validation steps that can lead to consolidation and broad adoption of this approach. Finally, the paper presents some future work that aiming at proving the approach.

International audience; We reviewed research papers related to pandemics/epidemics (disease outbreaks of a global/regional scope) published in major operations management, operations research, and management science journals through the end of 2019. We evaluate and categorize these papers. We study research trends, explore research gaps, and provide directions for more efficient and effective research in the future. In addition, our recommendations include the lessons learned from the ongoing pandemic, COVID-19. We discuss papers in the following categories: (a) warning signals/surveillance, (b) disease propagation leading to pandemic conditions, (c) mitigation, (d) vaccines and therapeutics development, (e) resource management, (f) supply chain configuration, (g) decision support systems for managing pandemics/epidemics, and (h) risk assessment.

International audience; The co-utilization of torrefied biomass and coal in thermochemical conversion technologies is an attractive process for the transition to green energy and chemicals. The main advantage of this process is the production of high quality oil products (closer resembling crude-oil) via synergistic interaction of primary products from the two feedstocks. These synergistic reactions often involve secondary reactions, which are promoted at high pressures. This paper reports quantitative results on the extent of synergistic reactions and the role of pressure on these reactions during co-pyrolysis of torrefied biomass and coal. Torrefied biomass was produced at 280 • C in a pilot rotary kiln and subsequently both pyrolysis and co-pyrolysis of torrefied biomass and coal was investigated in a fixed bed reactor (heating rate 7 • C/min) at temperatures of 400-600 • C and pressures of 1, 15 and 30 bar. The results show that the prior removal of hemicellulose during torrefaction maximized the potential of hydrogen transfer from cellulose/lignin-derived products to depolymerized coal fragments. Furthermore, the dehydration and condensation reactions of depolymerized fragments were suppressed during co-pyrolysis in favour of synergistic reactions between the fragments. These reactions occurred predominantly in the molten/ liquid phase and their rates could be indirectly controlled by pressure (through changing the evaporation rate), resulting in substantial changes in product distribution. In the presence of coal and its released vapours, the methoxyphenols (guaiacols) and furanics yields were significantly enhanced (positive deviation >54% and >40%, respectively, compared to additive predictions), whereas the phenol yields were inhibited (negative deviation >20%); suggesting the inhibition of demethoxylation reactions in the presence of coal depolymerized fragments. To date no studies have reported the molten phase synergistic reaction pathways occurring during copyrolysis of torrefied biomass and coal. Based on the findings in this paper, reaction pathways were proposed for these molten phase synergistic reactions.

Most public blockchain protocols, including the popular Bitcoin and Ethereum blockchains, do not formally specify the order in which miners should select transactions from the pool of pending (or uncommitted) transactions for inclusion in the blockchain. Over the years, informal conventions or "norms" for transaction ordering have, however, emerged via the use of shared software by miners, e.g., the GetBlockTemplate (GBT) mining protocol in Bitcoin Core. Today, a widely held view is that Bitcoin miners prioritize transactions based on their offered "transaction fee-per-byte." Bitcoin users are, consequently, encouraged to increase the fees to accelerate the commitment of their transactions, particularly during periods of congestion. In this paper, we audit the Bitcoin blockchain and present statistically significant evidence of mining pools deviating from the norms to accelerate the commitment of transactions for which they have (i) a selfish or vested interest, or (ii) received dark-fee payments via opaque (non-public) side-channels. As blockchains are increasingly being used as a record-keeping substrate for a variety of decentralized (financial technology) systems, our findings call for an urgent discussion on defining neutrality norms that miners must adhere to when ordering transactions in the chains. Finally, we make our data sets and scripts publicly available., This is a pre-print of our paper accepted to appear to ACM IMC 2021

International audience; Systems engineering, or engineering in general, has long been relying on document-centric approaches. Switching to model-based systems engineering, or MBSE for short, has extensively been discussed over the past three decades. Since about two decades, MBSE has been commonly associated with the modeling language SysML (Systems Modeling Language), which offers a standardized notation, not a methodology of using it. SysML needs therefore to be associated with a methodology supported by tools. In this paper, a methodology supported by the free and open-source software TTool is associated with SysML. This paper focuses discussion on methodological issues, leading the authors to share their experience in real-time systems modeling. Modeling with SysML is more than just drawing the different diagrams. Associated tools offer possibilities to analyze SysML models for specific properties. In this paper, verification addresses both safety and security properties. The TTool model checker inputs the SysML model enriched with safety properties to be verified and outputs an yes/no answer for each property. Security verification checks SysML models against confidentiality, integrity, and authenticity properties. As an illustration of the proposed approach, an aircraft cockpit door control system is modeled in SysML and verified against safety and security properties.

International audience; Business process optimization is a strategic activity in organizations because of its potential to increase profit margins and reduce operational costs. One of the main challenges in this activity is concerned with the problem of optimizing allocation and sharing of resources. Companies are continuously adjusting their resources to their needs following different strategies. However, the dynamic provisioning strategies are hard to compare. This paper proposes an automatic analysis technique to evaluate and compare the execution time and resource occupancy of a business process relative to a workload and a provisioning strategy. Four different strategies are presented, which are guided-respectively-by recent resource usage, recent resource request, predicted behavior, and a combination of available strategies. Such analysis is performed on models conforming to an extension of BPMN with quantitative information, including resource availability and constraints. Within this framework, the approach is fully mechanized using a formal and executable specification in the Maude rewriting logic framework, which relies on existing techniques and tools for simulating probabilistic and real-time specifications. The paper includes results on the extensive experimentation that has been carried out for validation purposes.

The use of two-phase nozzles for low-grade heat valorization by electricity production increases the energy recovery rate using Trilateral Flash or Wet to Dry cycles. A model benchmark for nozzle flow rate and efficiency estimation was conducted on experimental data from the literature. These data come from a series of tests made on geothermal energy production water two-phase nozzles. A new transitional bubble-to-droplet interfacial area density formulation (TA-BD model) is presented by the paper. It is compared to three models from literature. Two nozzles operating with different inlet and outlet conditions were modeled. The calibration flexibility and the robustness of the models are discussed in association with physical analysis. The paper shows how the models using single or redundant adaptation parameters fail to provide good results simultaneously on flow rate and efficiency. It appeared that the TA-BD model is the more flexible and robust. The Homogeneous Relaxation Model (HRM) model gives also good results. Furthermore, TA-BD model gives the lowest average discrepancies. Especially at the best efficiency point of the first test case, TA-BD model shows less than 1% discrepancy where the HRM model has 18% discrepancy in efficiency. The TA-BD model appeared to be easier to calibrate than the HRM model. Finally, regarding the proposed TA-BD model, the sensitivity to the geometry and operating conditions shows that the interfacial area density formulation could be completed to include the effect of nozzle's section profile, the effect of the inlet temperature on bubbles number density, and the effect of outlet pressure on the droplets number density.

International audience; Dwindling fossil fuel reserves and global climate change drive researchers to discover and develop new strategies to derive energy from renewable sources, such as biomass, including wood. However, when poorly controlled, wood burning can be a source of atmospheric pollution. With the ultimate purpose of better controlling pollutant emissions from domestic small combustion installations, this paper follows two objectives. First, temperature and pollutant measurements obtained at the chimney outlet of a domestic inset fed with wood logs and working under nominal operating conditions are presented. Measured pollutants include CO, CO 2 , nitrogen oxides (NOx), monoaromatic, carbonyl, and phenolic compounds, as well as polycyclic-aromatic hydrocarbons (PAHs) and sugars with measurement made using different types of wood. Second, this paper describes a first attempt of modeling based on a detailed chemistry and tests it for simulating the measured pollutants. The model includes a previously developed detailed chemical kinetic mechanism and a simplified model of thermal transfer in the wood log. A network of ideal reactors (equivalent reactor network (ERN)), for which simulations using the detailed kinetic model are feasible, is proposed to represent both the primary pyrolysis and the combustion of the emitted gaseous species. With only adjusting the parameters used in the model in order to simulate the smoke temperature and the CO 2 mole fraction for a single batch well, simulations give a reasonably good order of magnitude for all of the measured pollutants for the seven used wood batches. A sensitivity analysis of the used parameters and of the structure of the ERN model is also presented.

Time series are ubiquitous in data mining applications. Similar to other types of data, annotations can be challenging to acquire, thus preventing from training time series classification models. In this context, clustering methods can be an appropriate alternative as they create homogeneous groups allowing a better analysis of the data structure. Time series clustering has been investigated for many years and multiple approaches have already been proposed. Following the advent of deep learning in computer vision, researchers recently started to study the use of deep clustering to cluster time series data. The existing approaches mostly rely on representation learning (imported from computer vision), which consists of learning a representation of the data and performing the clustering task using this new representation. The goal of this paper is to provide a careful study and an experimental comparison of the existing literature on time series representation learning for deep clustering. In this paper, we went beyond the sole comparison of existing approaches and proposed to decompose deep clustering methods into three main components: (1) network architecture, (2) pretext loss, and (3) clustering loss. We evaluated all combinations of these components (totaling 300 different models) with the objective to study their relative influence on the clustering performance. We also experimentally compared the most efficient combinations we identified with existing non-deep clustering methods. Experiments were performed using the largest repository of time series datasets (the UCR/UEA archive) composed of 128 univariate and 30 multivariate datasets. Finally, we proposed an extension of the class activation maps method to the unsupervised case which allows to identify patterns providing highlights on how the network clustered the time series.

International audience; Nowadays, many research studies and industrial investigations have allowed the integration of the Internet of Things (IoT) in current and future networking applications by deploying a diversity of wireless-enabled devices ranging from smartphones, wearables, to sensors, drones, and connected vehicles. The growing number of IoT devices, the increasing complexity of IoT systems, and the large volume of generated data have made the monitoring and management of these networks extremely difficult. Numerous research papers have applied Reinforcement Learning (RL) and Deep Reinforcement Learning (DRL) techniques to overcome these difficulties by building IoT systems with effective and dynamic decision-making mechanisms, dealing with incomplete information related to their environments. The paper first reviews pre-existing surveys covering the application of RL and DRL techniques in IoT communication technologies and networking. The paper then analyzes the research papers that apply these techniques in wireless IoT to resolve issues related to routing, scheduling, resource allocation, dynamic spectrum access, energy, mobility, and caching. Finally, a discussion of the proposed approaches and their limits is followed by the identification of open issues to establish grounds for future research directions proposal.

In this paper, we analyze the effect of an enhanced voltage flip technique on the power performance of a piezoelectric energy harvester. The enhanced voltage flip principle is based on a synchronized-switch-based architecture, and is referred to as FAR (Full Active Rectifier). It uses a tiny amount of the stored charge to boost the voltage flip. This work aims to demonstrate that, beside the enhanced flip efficiency, the FAR also contributes to improve the power efficiency of the harvester, especially under changing load constraint. Therefore, the paper proposes a thorough comparison between the FAR and its conventional counterpart, the Switch-only technique. The FAR is easy to implement and does not require any external inductor or capacitor. It only needs a reduced set of switches, an active diode and a simple control sequence, and can thus be implemented on a fully integrated circuit. The FAR can be used as a standalone voltage flip solution or in addition to further boost the flip efficiency in a state-of-the-art architecture such as SSHC for example. Tests were performed on a 0.35-µm process CMOS prototype IC. Experimental results revealed that the FAR extracts 19.1μW from an off-the-shelf piezoelectric transducer when the output voltage is regulated at 1 V with 1 V open-circuit voltage and delivers up to 20% more power than the conventional Switch-only technique under load constraint. It also shows over 11× power efficiency improvement compared to a conventional diode-based full bridge rectifier.

An integrated dynamic model of natural gas pipeline networks is developed in this paper. Components for gas supply, e.g., pipelines, junctions, compressor stations, LNG terminals, regulation stations and gas storage facilities are included in the model. These components are firstly modeled with respect to their properties and functions and, then, integrated at the system level by Graph Theory. The model can be used for simulating the system response in different scenarios of operation, and evaluate the consequences from the perspectives of supply security and resilience. A case study is considered to evaluate the accuracy of the model by benchmarking its results against those from literature and the software Pipeline Studio. Finally, the model is applied on a relatively complex natural gas pipeline network and the results are analyzed in detail from the supply security and resilience points of view. The main contributions of the paper are: firstly, a novel model of a complex gas pipeline network is proposed as a dynamic state-space model at system level; a method, based on the dynamic model, is proposed to analyze the security and resilience of supply from a system perspective.

PurposeThe article provides a conceptual replication and enrichment of the status quo bias theory in the specific context of understanding IT department user resistance and user adoption. The findings can assist technology renewals and associated change management professionals to assess and plan the adoption and active usage of human resource systems.Design/methodology/approachThe authors used survey method to gather data. All items were based on prior literature. They administrated the survey to employees of GOODTECH (name changed), information systems (IS) department members, situated in France. They obtained 103 valid responses along with usage data from the system to run their path model, in order to validate the proposed research model.FindingsThe study offers an enriched user resistance model (URM) to understand why IT-savvy employees would resist or adopt new human resource tools. Apart from providing partial validity to status quo bias theory in the French context, the enriched model uses behavioral intention to use as an intermediate variable to explain the influence of two key constructs of the original theory: switching cost and switching benefits. This research provides a better explanatory power to understand the cause of user resistance and new IT use.Research limitations/implicationsThe sample size used in the study can be considered as a limitation, although power analysis reveals that the results are significant and valid. The context of the study is also limited to one country and to a specific type of IS implementation scenario. Since the purpose of the paper was to offer contextual theory enhancement, the findings are valid for this purpose.Practical implicationsDigital project managers are offered a framework to increase technology adoption of new human resource tools and evaluate how to reduce user resistance at times of technology renewals. Self-efficacy for change and colleagues’ opinion can indirectly impact behavioral intention to use via switching cost and switching benefit perceptions and thus reducing resistance perceptions as well as increasing adoption of new IT tools in post-implementation phases.Originality/valueThe paper enriches the well-established user resistance theory in IS domain in a context of human resource post-implementation phase by studying IT-savvy end user's perceptions. The paper demonstrates the need to integrate user adoption and user resistance variables in one parsimonious framework and extends support to emerging research on dual focus perspective.

In this paper, we propose a method for coarse camera pose computation which is robust to viewing conditions and does not require a detailed model of the scene. This method meets the growing need of easy deployment of robotics or augmented reality applications in any environments, especially those for which no accurate 3D model nor huge amount of ground truth data are available. It exploits the ability of deep learning techniques to reliably detect objects regardless of viewing conditions. Previous works have also shown that abstracting the geometry of a scene of objects by an ellipsoid cloud allows to compute the camera pose accurately enough for various application needs. Though promising, these approaches use the ellipses fitted to the detection bounding boxes as an approximation of the imaged objects. In this paper, we go one step further and propose a learning-based method which detects improved elliptic approximations of objects which are coherent with the 3D ellipsoid in terms of perspective projection. Experiments prove that the accuracy of the computed pose significantly increases thanks to our method and is more robust to the variability of the boundaries of the detection boxes. This is achieved with very little effort in terms of training data acquisition -- a few hundred calibrated images of which only three need manual object annotation. Code and models are released at https://github.com/zinsmatt/3D-Aware-Ellipses-for-Visual-Localization., Comment: Presented at 3DV 2020. Code and models released at https://github.com/zinsmatt/3D-Aware-Ellipses-for-Visual-Localization

International audience; Electricity plays a significant role in daily life and is the main component of countless applications. Thus, ongoing research is necessary to improve the existing approaches, or find new approaches, to enhancing power generation. The thermoelectric generator (TEG) is among the notable and widespread technologies used to produce electricity, and converts waste energy into electrical energy using the Seebeck effect. Due to the Seebeck effect, temperature change can be turned into electrical energy; hence, a TEG can be applied whenever there is a temperature difference. The present paper presents the theoretical background of the TEG, in addition to a comprehensive review of the TEG and its implementation in various fields. This paper also sheds light on the new technologies of the TEG and their related challenges. Notably, it was found that the TEG is efficient in hybrid heat recovery systems, such as the phase change material (PCM), heat pipe (HP), and proton exchange membrane (PEM), and the efficiency of the TEG has increased due to a set of improvements in the TEG’s materials. Moreover, results show that the TEG technology has been frequently applied in recent years, and all of the investigated papers agree that the TEG is a promising technology in power generation and heat recovery systems.

In this work, microwave planar resonators are printed with silver nanoparticle inks using two printing technologies, inkjet printing and aerosol jet printing, on polyimide substrates. The microwave resonators used in this paper operate in the frequency band 5–21 GHz. The printing parameters, such as the number of printed layers of silver nanoparticle inks, drop spacing, and sintering time, were optimized to ensure repeatable and conductive test patterns. To improve the electrical conductivity of silver deposits, which are first dried using a hot plate or an oven, two complementary sintering methods are used: intense pulsed light (IPL) and laser sintering. This paper presents the results of different strategies for increasing the final quality factor of printed planar resonators and the trade-offs (sintering time versus final conductivity/unloaded Q) that can be reached. Improvement of the resonator unloaded quality factor (up to +55%) and of the equivalent electrical conductivity (up to 14.94 S/μm) at 14 GHz have been obtained thanks to these nonconventional sintering techniques. The total sintering durations of different combinations of sintering techniques (hot plate, oven, IPL, and laser) range from 960 to 90 min with a final conductivity from 14.94 to 7.1 S/μm at 14 GHz, respectively.

The Two-Dimensional Fast Fourier Transform (2D-FFT) algorithm is used for the study of many modern systems applied for security and biometrics. The adoption of this algorithm, which is a compute intensive task, is limited due to its hardware design complexity. The first objective of this paper is to underline the effect of the hardware/software co-design (Hw/Sw co-design) for the reduction of the processing time and power consumption. Secondly, we propose an innovative architecture for the 2D-FFT algorithm tested on Zynq Soc, which requires less processing time and memory compared to the traditional algorithm. Three implementations (one software and two Hw/Sw co-designs) of the 2D-FFT algorithm using the Zynq SoC are presented in this paper. The first is based on ARM processor. A speedup of 29x is obtained compared to the original implementation thanks to many optimizations. The second is a Hw/Sw co-design solution of the traditional 2D-FFT algorithm introduced on a hybrid platform combining an ARM Cortex-A9 processor with an FPGA. The third is also a Hw/Sw co-design solution using our optimized 2D-FFT algorithm to reach the real-time contraints for high-resolution images (1920 × 1080). It provides a speedup of 1.13x, 3.31x and 96.21x faster than the Hw/Sw co-design implementation of the traditional RC algorithm, the pure software implementations with and without optimizations, respectively.

The Last decade saw significant growth in the digital terrestrial television (DTT) subscribers. The High peak-to-average power ratio (PAPR) of the transmitted signal is implied by the large fluctuations in its amplitude, which is a serious issue in the contemporary and next generation DTT broadcasting systems. These large fluctuations prevent feeding the high power amplifier (HPA) at an operating point near its non-linear (NL) saturation region, thereby lowering its power efficiency. For a better future, energy-efficient green communications is the need of hour. This paper focuses on digital predistortion (DPD), which is among the widely used techniques for HPA linearization in order to minimize the NL effects. DPD technique requires a feedback signal. This paper proposes a novel adding signal-based digital predistortion, where the output of the HPA is gradually linearized by adding a signal at the input in an iterative manner. By Taylor series approximation, the DPD problem can be posed as a convex one. We prove that a tighter upper bound exists for the convergence factor and instead of directly solving the convex problem, we pick a convergence factor value within its bounds that offers a good trade-off between convergence speed and residual error value. The simulation results show that the proposed technique can quickly linearize the HPA with good convergence.

International audience; The Internet of Things (IoT) is currently used in many wireless sensor network (WSN) applications. Conventionally, the WSN’s energy consumption is considered as one of the prime concerns. The energy consumption is mostly due to the sensing, data processing, communications and other wasted energies such as idle listening, collisions and overhearing. These sensor nodes are regularly powered by an external power source and consequently have a shorter lifetime. Fortunately, different approaches for wireless energy harvesting (WEH) have been improved. Thus, wake-up radio (WuR) becomes a remedy for WEH which provides active, passive and semi-passive circuits consumption and other protocols. For instance, the most used, convincing and productive go to passive WuR which can significantly increase the network lifetime in a sensor network (SN) by decreasing unnecessary idle listening. Eventually, this paper proposes a state of the art in active, semi-passive, centering more on passive WuR and covers the applications areas. Then, an overview of WuR related to physical, medium access control (MAC) and routing layers. Lastly, the paper accentuates the potential research opportunities for wake-up technology for future IoT applications.

International audience; Copy-move is a very popular image falsification where a semantically coherent part of the image, the source area, is copied and pasted at another position within the same image as the so-called target area. The majority of existing copy-move detectors search for matching areas and thus identify the source and target zones indifferently, while only the target really represents a tampered area. To the best of our knowledge, at the moment of preparing this paper there has been only one published method called BusterNet that is capable of performing source and target disambiguation by using a specifically designed deep neural network. Different from the deep-learning-based BusterNet method, we propose in this paper a source and target disentangling approach based on local statistical model of image patches. Our proposed method acts as a secondstage detector after a first stage of copy-move detection of duplicated areas. We had the following intuition: even if no manipulation (e.g., scaling and rotation) is added on target area, its boundaries should expose a statistical deviation from the pristine area and the source area; further, if the target area is manipulated, the deviation should appear not only on the boundaries but on the full zone. Our method relies on machine learning tool with Gaussian Mixture Model to describe likelihood of image patches. Likelihoods are then compared between the pristine region and the candidate source/target areas as identified by the first-stage detector. Experiments and comparisons demonstrate the effectiveness of the proposed method.

The current evolution of technical and environmental regulations and the recent development of hydrogen mobility require technological breakthroughs in the industrial areas, such as in harbours. Thus, multi-energy microgrids integrating low-carbon energy sources, electrical and hydrogen loads and storage solutions have to be designed. To improve their economic viability, the microgrid components must be carefully sized and the energy must be distributed in the most cost-effective way at any time. So as to foster the expansion of multi-energy microgrids in industrial areas, we propose in this paper a two-level optimization for the energy management and the sizing, applied to an original multi-energy scenario considering electricity and hydrogen as energy vectors. The designed energy management optimization takes into account an objective of economic profitability and constraints related to the availability of the sources and storage solutions, their reliability and the costs. The sizing optimization aims to propose different solutions allowing the benefits to be maximized and the expenses to be minimized. One of the main contributions of the paper is to compare the possible ways of valorisation of the energy available in an industrial microgrid. The results show that the sale of hydrogen allows income to be increased, in addition to self-consumption and sale to the electrical grid, but the electrolyzer involves high investments costs. Finally, a sensitivity analysis is presented and shows that the investment costs of electrolyzer and battery and the hydrogen selling price are key points in the optimal design.

This paper uses the fuzzy theory basics to assess accident risk that may damage a target (e.g., humans, assets, or the environment). Converting a target’s position to a membership degree in a fuzzy danger zone (FDZ) helps calculate risk indices. Using FDZs normalizes the effects of different kinds of hazards, similarly visualizes them, and distinguishes the impact of a threat on various types of targets. This paper presents a related mathematical formulation. The proposed approach evaluates the accident risks by simulating industrial activities in CAD, virtual reality, and augmented reality when using a concurrent engineering platform. This approach can calculate the risk index during a human task simulation and through real-time human interaction with a virtual machine during safety analysis and training. The results also are credible to activate alarm systems according to the operator’s limbs place during work. The model provides a normalized and similar scale for various risks (e.g., biological, chemical, and physical hazards) and computes the effect of danger on different target types.

URSI GASS 2021, XXXIV General Assembly and Scientific Symposium (GASS) of the International Union of Radio Science (URSI), Rome, ITALIE, 28-/08/2021 - 04/09/2021; International audience; In the context of the digital transformation of railway transportation systems, railway operators investigate the potentialities of the IoT (Internet of Things) in order to optimize the maintenance operations on the vehicles and the infrastructures. One of the interest is to use IoT communication solutions in order to centralize the data measured by the different sensors distributed everywhere along the railway infrastructures and on board the trains. In this perspective, different communication protocols could be employed. This paper deals specifically with the LoRa communication protocol and analyses its susceptibility in the presence of railway electromagnetic interferences. This paper presents a susceptibility analysis of the LoRa communications in the presence of transient EM interferences representative of the electromagnetic interferences produced by the sliding catenary-pantograph contact.

conférence virtuelle; International audience; This paper presents a blockchain cloud deployment for an industrial vehicle use case. Hyperledger Sawtooth is used as distributed ledger with Rancher and Kubernetes as container-orchestration. This paper aims to explore the feasibility and scalability of the use case in a real-world scenario cloud deployment. Also, the blockchain performances is analyzed by stressing the implementation with virtual IoT clients, changing the blockchain settings, number of peer members and editing the core software default parameters. Benchmarks will reveal that Hyperledger Sawtooth performance can be as high as 25 transaction per seconds for a network of 4 nodes with PBFT consensus. Results lead us to discuss the possible components reducing throughput and blockchain commit rate.

This paper received the Best Paper Award for IWCC 2021; International audience; Malware is a primary concern in cybersecurity, being one of the attacker's favorite cyberweapons. Over time, malware evolves not only in complexity but also in diversity and quantity. Malware analysis automation is thus crucial. In this paper we present ECDGs, a shorter call graph representation, and a new similarity function that is accurate and robust. Toward this goal, we revisit some principles of malware analysis research to define basic primitives and an evaluation paradigm addressed for the setup of more reliable experiments. Our benchmark shows that our similarity function is very efficient in practice, achieving speedup rates of 3.30x and 354, 11x wrt. radiff2 for the standard and the cache-enhanced implementations, respectively. Our evaluations generate clusters that produce almost unerring results-homogeneity score of 0.983 for the accuracy phase-and marginal information loss for a highly polluted dataset-NMI score of 0.974 between initial and final clusters of the robustness phase. Overall, ECDGs and our similarity function enable autonomous frameworks for malware search and clustering that can assist human-based analysis or improve classification models for malware analysis.

The smart grid systems allow power and data flows between suppliers and consumers to optimize the grid in terms of sustainability, reliability, and efficiency. The estimate of time-based energy consumption and generation plays a crucial role in future energy systems. The electric generation systems are changing fast. The complexity of electric power systems has been growing because of renewables, inverter-based control devices and integration of electrical vehicles. The most important challenge in a smart grid is how to reduce the cost of power with a large amount of user participation. The developing smart grid featured by monitoring, security and renewable energy demand-generation control is facing difficulties because of enormously large data sets. Different types of faults have occurred in smart grids that cause major maintenance costs. Effective monitoring tools and predictive approaches can help to maintain a reliable supply of electric energy. Based on the large data sets collected by measurement devices, many data analytics methods are applied in the smart grid to solve various problems. This paper presents a fault prediction approach and load forecasting approach. The approach of data analytics and deep learning for smart grids (SGs) and their applications are proposed in this paper, which seeks to provide a transition to future energy systems while providing significant knowledge of existing smart grids.

Output active and reactive powers of battery chargers of electric vehicles are controlled for participation in vehicle-to-grid technology. In this paper, a typical control system, suitable for vehicle-to-grid technology, is equipped with third harmonic injection technique. This paper aims to evaluate how much third harmonic injection changes level of the output active and reactive powers as well as the dynamic response. Results consider two scenarios for pulse-width modulation, in the presence and in the absence of third harmonic injection, and reveal that it is a highly effective method to enhance the level of the output reactive power and the speed of change of the output active power.

E-Agriculture, or Smart Farming, refers to the design, development, and application of innovative methods to use modern information and communication technologies (ICTs), such as the Internet of Things (IoT) and machine learning, to move towards more sustainable agricultural and farming practices. The integration of blockchain technology in farming is gaining attention for its potential to migrate from the centralized and monopolistic model that shapes today’s food value chain. This paper highlights the fact that most of today’s blockchain-based farming frameworks focus on food tracking and traceability. Only rarely does research focus on the design of digital marketplaces to support the trading of agricultural goods between farmers and potentially interested third party stakeholders; equally rarely are performance evaluations performed for the proposed frameworks. The latter is where this paper contributes the most by, not only proposing a novel blockchain-based farming marketplace platform (called “FarMarketplace”), but also a comprehensive methodology to help software solution integrators to better understand and measure how a given configuration setting of such a platform can influence the overall quality of service performance in the long run.

International audience; The rapid explosion of linked data has resulted into many weakly structured and incomplete data sources, where typing information might be missing. On the other hand, type information is essential for a number of tasks such as query answering, integration, summarization and partitioning. Existing approaches for type discovery, either completely ignore type declarations available in the dataset (implicit type discovery approaches), or rely only on existing types, in order to complement them (explicit type enrichment approaches). Implicit type discovery approaches are based on instance grouping, which requires an exhaustive comparison between the instances. This process is expensive and not incremental. Explicit type enrichment approaches on the other hand, are not able to identify new types and they can not process data sources that have little or no schema information. In this paper, we present HInT, the first incremental and hybrid type discovery system for RDF datasets, enabling type discovery in datasets where type declarations are missing. To achieve this goal, we incrementally identify the patterns of the various instances, we index and then group them to identify the types. During the processing of an instance, our approach exploits its type information, if available, to improve the quality of the discovered types by guiding the classification of the new instance in the correct group and by refining the groups already built. We analytically and experimentally show that our approach dominates in terms of efficiency, competitors from both worlds, implicit type discovery and explicit type enrichment while outperforming them in most of the cases in terms of quality.

International audience; Decision lists (DLs) find a wide range of uses for classification problems in Machine Learning (ML), being implemented in anumber of ML frameworks. DLs are often perceived as interpretable. However, building on recent results for decision trees (DTs), we argue that interpretability is an elusive goal for some DLs. As a result, for some uses of DLs, it will be important to compute (rigorous) explanations. Unfortunately, and in clear contrast with the case of DTs, this paper shows that computing explanations for DLs is computationally hard. Motivated by this result, the paper proposes propositional encodings for computing abductive explanations (AXps) and contrastive explanations (CXps) of DLs. Furthermore, the paper investigates the practical efficiency of a MARCO-like approach for enumerating explanations. The experimental results demonstrate that, for DLs used in practical settings, the use of SAT oracles offers a very efficient solution, and that complete enumeration of explanations is most often feasible.

International audience; One of the cornerstones of the cloud provider business is to reduce hardware resources cost by maximizing their utilization. This is done through smartly sharing processor, memory, network and storage, while fully satisfying SLOs negotiated with customers. For the storage part, while SSDs are increasingly deployed in data centers mainly for their performance and energy efficiency, their internal mechanisms may cause a dramatic SLO violation. In effect, we measured that I/O interference may induce a 10x performance drop. We are building a framework based on autonomic computing which aims to achieve intelligent container placement on storage systems by preventing bad I/O interference scenarios. One prerequisite to such a framework is to design SSD performance models that take into account interactions between running processes/containers, the operating system and the SSD. These interactions are complex. In this paper, we investigate the use of machine learning for building such models in a container based Cloud environment. We have investigated five popular machine learning algorithms along with six different I/O intensive applications and benchmarks. We analyzed the prediction accuracy, the learning curve, the feature importance and the training time of the tested algorithms on four different SSD models. Beyond describing modeling component of our framework, this paper aims to provide insights for cloud providers to implement SLO compliant container placement algorithms on SSDs. Our machine learning-based framework succeeded in modeling I/O interference with a median Normalized Root-Mean-Square Error (NRMSE) of 2.5%.

International audience; Identifying a print document (original) from a reprint document (copy or fake) can be a challenge. The analyse at microscopic scale of print documents shows random dot shapes which depend on the printing parameters as well as the printing device used. We can, therefore, draw the assumption that the dot shapes can be used as a fingerprint to differentiate a print from a reprint. In this paper, we explore several shape indexes that were not investigated until now to analyse at microscopic scale documents printed on aluminium foils using rotogravure printing process. This paper presents a statistical analysis which is based on a pattern recognition process defined by three steps. First, a new image processing pipeline is used to segment automatically disconnected dots. Next, new dot pattern features are used to characterize automatically dot patterns. Six types of dot patterns (including four types of doughnut patterns) are introduced. Lastly, a new statistical analysis method is used to characterize a printed sample from the set of dots printed on it. The experiments done demonstrate the relevance of the analytical method proposed. Results shows the potential of this method to identify a reprint from a print.

International audience; Blockchain technology enables several untrustworthy parties to execute inter-organizational business processes in a tamper-proof manner. Existing approaches are based on smart contract code generation. Although the smart contract immutability guarantees that the process model is followed by each participant, any change in the process model makes the existing process instance smart contract not compliant with the process model and therefore, the deployment of a new smart contract becomes a requirement. In this paper, we propose CoBuP, a decentralized Collaborative Business Process execution architecture using blockchain. To address the inflexibility issue, this paper presents an interpreter of BPMN process models, supporting the instantiating, execution, monitoring of process instances. For this reason, one generic smart contract is deployed once, which helps generate a process instance, create BPMN elements by invoking the process instance predefined functions, and dynamically update the adaptation of a process at run time. For validation purposes, two use cases have permitted to carry out different experiments that demonstrate the technical feasibility of our approach. The obtained results of the experimental evaluation shows that our approach cost is similar or lower than that of the existing approaches.

Part 18: Smart Blockchain Applications/Cybersecurity; International audience; This paper presents a Validation and Verification (V&V) model of Data Marketplaces. Data is extracted from the sensors embedded within the Smart city, infrastructure, or building via Application Programming Interfaces (APIs) and inserted into a Data Marketplace. The technology is based on smart contracts deployed on a private ethereum blockchain. Current issues with data in Smart cities, infrastructure, buildings, or any real estate, are the difficulty of its access and retrieval, therefore integration; its quality in terms of meaningful information; its large quantity with a reduced coverage in terms of systems and finally its authenticity, as data can be manipulated for economic advantage. In order to address these issues, this paper proposes a Data Marketplace model with a hierarchical process for data validation and verification where each stage adds a layer of data abstraction, value-added services and authenticity based on Artificial Intelligence. By using a blockchain, this presented approach is based on a decentralised method where each stakeholder stores the data. The proposed model is validated in a real application with live data: Newcastle urban observatory smart city project.

Part 8: Data Mining/Word Counts; International audience; Word clouds are a very useful tool for summarizing textual information. They can be used to illustrate the most frequent and important words of text documents or a set of text documents. In that respect they can also be used for topic visualisation. In this paper we present an experiment investigating how the crowd understands topics visualised via word clouds. In the experiment we use the topics mined from Instagram hashtags of a set of Instagram images corresponding to 30 different subjects. By subject we mean the research hashtag we use to gather pairs of Instagram images and hashtags. With the aid of an innovative topic modelling method, developed in a previous work, we constructed word clouds for the visualisation of each topic. Then we used a popular crowdsourcing platform (Appen) to let users identify the topic they believe each word cloud represents. The results show some interesting variations across subjects which are analysed and discussed in detail throughout the paper. Given that the topics were mined from Instagram hashtags, the current study provides useful insights regarding the appropriateness of hashstags as image annotation tags.

Part 3: AI Impacts/Big Data; International audience; Artificial Intelligence brings exciting innovations in all aspects of life and creates new opportunities across industry sectors. At the same time, it raises significant questions in terms of trust, ethics, and accountability. This paper offers an introduction to the AI4Media project, which aims to build on recent advances of AI in order to offer innovative tools to the media sector. AI4Media unifies the fragmented landscape of media-related AI technologies by investigating new learning paradigms and distributed AI, exploring issues of AI explainability, robustness and privacy, examining AI techniques for content analysis, and exploiting AI to address major societal challenges. In this paper, we focus on our vision of how such AI technologies can reshape the media sector, by discussing seven industrial use cases that range from combating disinformation in social media and supporting journalists for news story creation, to high quality video production, game design, and artistic co-creation. For each of these use cases, we highlight the present challenges and needs, and explain how they can be efficiently addressed by using innovative AI-driven solutions.

Part 9: Deep Learning; International audience; Nowadays Internet of Things (IoT) is a major paradigm shift that will mark an epoch in communication technology such that every physical object can be connected to the Internet. 5G makes a significant breakthrough in the traditional mobile communication system and support the applications of IoT in various fields including business, manufacturing, health care and transportation.5G is increasing the service capability of future IoT and operates and connects the whole society. 5G is facing enormous challenges when it supports differentiated applications with a uniform technical framework. In recent years, Artificial Intelligence (AI) is rising to these challenges with the rapid development. It is a potential solution to the problems in the 5G era and will lead to a revolution in the capabilities and concepts of the communication systems. Many researches have already been done for applying AI in 5G. In this paper, we focus on clarifying the promising research directions with the greatest potential rather than trying to review all the existing literatures. In this research, 5G can be anticipated to achieve significantly better performance and more convenient implementations compared to the traditional communication systems. With the inspiring research paradigms introduced in this paper, we are looking forward to the remarkable achievements of AI in 5G in the near future.

Part 15: Natural Language; International audience; Keyphrase extraction is a fundamental task in information management, which is often used as a preliminary step in various information retrieval and natural language processing tasks. The main contribution of this paper lies in providing a comparative assessment of prominent multilingual unsupervised keyphrase extraction methods that build on statistical (RAKE, YAKE), graph-based (TextRank, SingleRank) and deep learning (KeyBERT) methods. For the experimentations reported in this paper, we employ well-known datasets designed for keyphrase extraction from five different natural languages (English, French, Spanish, Portuguese and Polish). We use the F1 score and a partial match evaluation framework, aiming to investigate whether the number of terms of the documents and the language of each dataset affect the accuracy of the selected methods. Our experimental results reveal a set of insights about the suitability of the selected methods in texts of different sizes, as well as the performance of these methods in datasets of different languages.

International audience; The paper deals with the monolithic integration of a power converter phase leg. It focuses on the integration of a power phase leg consisting of a VDMOS and an IGBT. This association is suitable for monolithic power integration in silicon. Different monolithic chips integrating phase legs were proposed and discussed in this paper. Interesting advantages can be brought by the monolithic integration of the converter phase leg. These advantages include power chips realization simplification, control of gate transistors with respect to constant voltages (DC-link or ground-0V).

International audience; This paper presents detectEr, a runtime verification tool for monitoring asynchronous component systems. The tool synthesises executable monitors from properties expressed in terms of the safety fragment of the modal μ-calculus. In this paper, we show how a number of useful properties can be flexibly runtime verified via the three forms of instrumentation-inline, outline, and offline-offered by detectEr to cater for specific system setup constraints.

International audience; As it had already been observed in other domains, such as aircraft or automotive sectors, the development of fully automated driving systems in the mainline railway sector faces multiple technical and safety barriers. Before such an advanced system can be overcome, systems with intermediate and adaptive levels of automation must be considered and studied. In this context, human operators maintain a crucial role for the train driving activity and their interactions with technical systems and other agents must remain among the main focuses of the conception phase. These interactions are considered through a thorough Human-Machine cooperation model. The objective of this paper is to present a conception and evaluation method implementing this model in the development of assistance systems for cooperation with human operators. The method is currently being applied for train remote driving as part of the TC-Rail project, where the aim of the study is to focus on Human-Factors aspects. A second phase of development is currently in progress and has been complemented by the results of a first phase. Along with the conception method, the state of this second phase, as well as use cases for future tests that implement it, are presented in this paper.