The design of learning games for learning is a complex task. It involves a large number of challenges for the different stakeholders (e.g. institutions, teachers, technical designers, players, video game experts). Among these challenges, we can note the acculturation to the game, the difficulty to align pedagogical concepts with the game mechanics and diegesis, or the specific needs of communities of practice. Consequently, we observe in the TEL community a strong ad hoc aspect of the design of serious games, especially regarding the game elements used to address specific pedagogical intentions. However, this ad hoc character does not allow to capitalize efficiently on both the serious games created, nor the choices between pedagogical intentions and game elements to implement them. The expertise of the whole community is then difficult to share and to reuse, and it is difficult to efficiently assist the actors in this design stage. The goal of the TALE4GDA project is to bring new assistance to the stakeholders in the design of learning games and to allow the capitalization of these experiences. To do so, we will propose a first formalization of the concept of alignment between a game entity and a pedagogical intention - a pedago-ludic alignment. This will allow us to propose the first topology of shareable alignments: each alignment will be characterized by its relations with the others (e.g. proximity, overlap). We will take a pioneering approach by allowing the annotation of these alignments in a controlled way, exploring even the possibility of exemplifying them with real situations. Thanks to this, we will be able to set up innovative mechanisms for decision support, design and capitalization based on automatic semantic and topological reasoning.

DEPOSIA focuses on the detection and geolocation of various radio frequency signal sources in order to thwart attacks on connected systems and infrastructures. The sources considered are elements which by their characteristics or their position, present an illicit character and which threaten the people security or the infrastructures. For outdoor cases, we consider drones flying over forbidden areas, telecommunication jammers, spoofing signal transmitters or wireless connected sensors used to introduce false data in monitoring platforms. For indoor cases, we also consider jamming or spoofing sources that can cause denial of service within networks or infrastructures, or fake access points that aim to carry out man-in-the-middle attacks to intercept information. In this proposal, the indoor and outdoor use cases are considered separately in order to design monitoring infrastructures adapted to each case. For the outdoor case, we consider a surveillance architecture that could join the already existing cellular or WLAN communication infrastructures. In particular, with 5G technology and the higher employed frequencies, cellular networks are evolving towards finer meshes and have interfaces with the core network at each of their nodes. Thus, these interface points, equipped with receivers dedicated to monitoring, could enable the routing of monitoring data to centralized platforms, feeding an Artificial Intelligence for analysis, anomaly detection and source geolocation. For the indoor case, we consider a distributed monitoring architecture deployed within a building, based on SDR sensors and a data centralization and synchronization network. In these two cases, we envisage an Artificial Intelligence working on data evolving in three dimensions : time, space and direction, all for data of different natures, namely those from the physical layer and the data link layer. Whether for indoor or outdoor configurations, the algorithms that will constitute the Artificial Intelligence will be based on learning approaches that will correspond to Machine Learning and Deep Learning algorithms. These algorithms will deal with the problems of detecting attacks and locating illicit sources. These algorithms will have to take into account: the evolutionary aspect brought by the non-fixed character in time of the attacks and the non-fixed location aspect of the localization of the source of the attack. A first Artificial Intelligence will be dedicated to data analysis and anomaly detection, i.e., highlighting the suspicious nature of the data, and a second Artificial Intelligence will be dedicated to extracting the location information of the attack source. Due to the multi-layered nature of the data, model aggregation algorithms will be deployed in order to homogenize the decision process.

Within the context of energy shift towards a decrease in the contribution of fossil fuels, the development of new stationary energy storage systems is mandatory. Indeed, the intrinsic intermittent and variable nature of renewable energy sources, such as windmill and photovoltaic, require energy storage. Redox-flow batteries, allowing a decoupling of energy and power, are well adapted to such requirements. As a matter of fact, this technology presents advantages as compared to Li-ion systems presently under development for such applications, in particular for security and recyclability issues. However, the most advanced redox-flow batteries (Vanadium redox-flow batteries, studied since the 80’s) remain expensive with limitations in terms of stability and capacities. The present project aims at developing a full redox-flow battery, based on the flow of redox-mediators based aqueous solutions (pH around 7), using sodium insertion materials immobilized in the battery tanks. The use of these insertion materials will allow an increase in the energy density of these systems, and thus to potentially reduce their size. These materials will be free of toxic or expensive metallic element. To perform these research studies, we created a multidisciplinary team which will allow to break the technological locks related to the development of such innovative and performing systems. The project partners will pursue in particular the study and development of a pilot battery so as to demonstrate the potentialities of this approach for electrochemical energy storage at large scale (coupling with windmill and photovoltaic systems), with storage time of the order of a dozen hours.

The XPM project aims to integrate explanations into Artificial Intelligence (AI) solutions within the area of Predictive Maintenance (PM). Real-world applications of PM are increasingly complex, with intricate interactions of many components. AI solutions are a very popular technique in this domain, and especially the black-box models based on deep learning approaches are showing very promising results in terms of predictive accuracy and capability of modelling complex systems. However, the decisions made by these black-box models are often difficult for human experts to understand – and therefore to act upon. The complete repair plan and maintenance actions that must be performed based on the detected symptoms of damage and wear often require complex reasoning and planning processes, involving many actors and balancing different priorities. It is not realistic to expect this complete solution to be created automatically – there is too much context that needs to be taken into account. Therefore, operators, technicians and managers require insights to understand what is happening, why it is happening, and how to react. Today’s mostly black-box AI does not provide these insights, nor does it support experts in making maintenance decisions based on the deviations it detects. The effectiveness of the PM system depends much less on the accuracy of the alarms the AI raises than on the relevancy of the actions operators perform based on these alarms. In the XPM project, we will develop several different types of explanations (anything from visual analytics through prototypical examples to deductive argumentative systems) and demonstrate their usefulness in four selected case studies: electric vehicles, metro trains, steel plant and wind farms. In each of them, we will demonstrate how the right explanations of decisions made by AI systems lead to better results across several dimensions, including identifying the component or part of the process where the problem has occurred; understanding the severity and future consequences of detected deviations; choosing the optimal repair and maintenance plan from several alternatives created based on different priorities, and understanding the reasons why the problem has occurred in the first place as a way to improve system design for the future.

Shipping is an essential transport infrastructure, with 80% of our goods undergoing overseas transport. However shipping emissions have impacts on climate change and on air quality, through the emission of gaseous (SO2, NOx, CO2, VOCs…) and particulate (PM) pollutants, particularly important for highly populated coastal areas. Since the 90s, regulations for emissions started to evolve, leading to the current limitations of fuel sulphur content (0.5%) and the application of Tier I - III standards for emissions. It is however likely than further changes need to be implemented to move towards more sustainable practices, particularly in harbors. But it is currently highly challenging to estimate the impact of shipping emissions on urban air quality, due, amongst others, to the transient nature of shipping plumes, the differences between vessels and fuels used, and the lack of understanding of the chemical evolution of the pollutants, which currently hamper accurate modelling of current and future changes. The project SHIPAIR therefore proposes to tackle some of these challenges through an interdisciplinary approach, combining online (& off-line) measurements with real-time shipping data through the automatic identification system and novel modelling approaches. Two field campaigns will inform not only on the pollutants emitted, but also on their evolution during transport and on their oxidation potential (OP). The first measurement campaign is an intensive (3 weeks) field campaign in the harbor of Dunkirk. Measurements on two locations, one near-field and one close to the urban border, of a comprehensive suite of pollutants (gas and particulates, including on-line metal speciation) will allow a better estimate of their evolution, their influence on the OP and on urban AQ. Furthermore, the deployment of a photochemical flow reactor will allow to assess the secondary aerosol formation potential of the plumes. The second measurement campaign will take place during one year in an urban monitoring site in Marseille, focusing on the deconvolution of different source contributions, in particular to the OP. The deployment for the first time of a novel online instrument to measure OP with a 20-minute time resolution over a long time period (3-4 months), will produce a unique, high resolution data set. The data obtained through these campaigns will be analyzed using state-of-the-art positive matrix factorization (PMF) in order to disentangle different source contributions. For the local AQ networks (AASQA) involved in SHIPAIR, a major challenge in predicting AQ in port cities, is the unequal access and use of information. To counter this difficulty, the 3 AASQAs will work closely together to harmonize and standardize their modelling approach in close collaboration with the port authorities. The emission inventories used will be enlarged based on literature and ongoing projects. Another difficulty in modelling the AQ of urban center close to harbors, lays in the resolutions of the models and their limited representation of atmospheric processes, affecting notably the accuracy of prediction for ultra-fine particles and the chemical composition of PM. SHIPAIR proposes to develop a new dispersion modelling framework for ship plumes in urban areas, based on a “plume-in-grid” and a “street-in-grid” approach. Furthermore, the model will integrate the treatment for metallic compounds in the SSH-aerosol module, allowing to investigate the contribution of metals to the OP. This new modelling framework will be evaluated against the measurement dataset from the campaigns and the AASQAs. Finally, SHIPAIR will compare the impact of shipping emissions determined by the different methodologies (PMF and model with and without shipping emission) for different harbors (Dunkirk, Marseille and Le Havre). After validation first runs of scenarios for future trends will be implemented by each AASQA to evaluate the impact of local mitigation strategies.