Cynotechnicians are involved in the security of places open to the public. Their effectiveness depends largely on the quality of communication between the handler and the dog and on the discretion of this communication based on voice and gestures. The necessary presence of the handler limits the effective availability of the dog for its security mission. Even if the dog is available, the handler may be absent. Communication based on vibrations delivered by means of an instrumented harness may be an interesting solution. The HAPNESS project aims to design and explore the use of vibration interaction (also known as haptic interaction) for surveillance applications where dogs can receive commands not exclusively from their handler. The HAPNESS project will study the sensitivity of dogs to haptic interaction and investigate the training required for the dog to learn haptic patterns. HAPNESS will be based on the design of a harness equipped with vibratory actuators combined with sensors to monitor the dog’s behavior during its mission. Communication will be handled wirelessly with a touchpad. The evaluation will take place on an operational scenario of doubt removal and will study the efficiency of this type of communication when the orders are delivered by the main handler of the dog and by another handler.

Machine learning methods have become prevalent in language technologies. They rely on annotated corpora to train and evaluate models. The CoDeinE project proposes to address the lack of shareable corpora in sensitive domains such as health or banking. The key idea of the project is to define methods for paraphrase generation and apply them to confidential corpora to automatically generate synthetic texts that mimic the linguistic properties of real documents while preserving confidentiality. The project addresses important issues in natural language processing and is also concerned with defining confidentiality criteria to ensure that no original confidential information is found in the generated synthetic texts. We will use clinical documents in electronic patient records as a case study. Furthermore, the project will rely on Games With A Purpose and crowd sourcing to validate and annotate the synthesized texts.

About 466 million people worldwide suffer from hearing loss. Of these, 34 million live in the EU and 6 million in France. With the aging of the world's population, this number is expected to increase to 900 million by 2050, WHO says. Still according to the WHO, disabling hearing loss can lead to depression, loneliness and social isolation, as well as reduced levels of employment and income. The cost associated with hearing loss is currently estimated to €216 billion per year in Europe (Hear-it, 2019) due to reduced productivity, reduced quality of life, and health and societal costs. Current estimates suggest that in France (Kervasdoué-Hartmann report 2016), only 30-35% people who would need a hearing aid actually use one. Current hearing aids provide a poor experience in noisy environments or for multiple sound streams, which is detrimental to social communication. In addition, patients must return to the hearing care professional sometimes more than ten times a year (at the beginning) to adjust settings. To improve the efficiency of hearing aids, one solution is to equip them with filtering/separation algorithms to isolate the relevant streams. Machine learning has made the use of these algorithms credible in real life conditions, in often-complex scenarios. However, current artificial intelligence methods are too complex to be applied to these portable devices equipped with processors with low computing and memory capacities. Moreover, existing filtering/separation algorithms are generally not adapted to the particular characteristics of the patients' hearing loss. The REFINED project is based on the upstream identification of auditory and extra-auditory spectro-temporal cues that correlate with the level of speech perception in people with auditory neuropathy spectrum disorders, constituting the 10% fringe of subjects who are more in need of speech filtering/separation than the primary function of conventional hearing aids (sound amplification). We study and develop efficient algorithms for auditory stream separation based on machine learning. We simplify them in order to implement them in real time while maintaining their performance and test their effectiveness on a cohort of carefully selected volunteer patients. This is done in an original approach in constant interaction between algorithms, embedded development and tests on patients. We develop experimental strategies based on the knowledge of sound perception to tune the algorithms in order to optimize speech recognition with respect to the limitation of information transfer and processing. The consortium brings together partners from three very different backgrounds, who will pool their respective expertise to design a system that naturally lies at the border of their three worlds. The Institute of Hearing (Pasteur Institute) ensures the selection of subjects with well-defined audiological profiles on whom to test speech-processing systems, the LORIA (University of Nancy) brings skills in machine learning applied to speech enhancement and sound source separation and the CEA (project coordinator) brings its skills in artificial intelligence and design of performance constrained embedded systems. To the best of our knowledge, REFINED will be the first initiative aiming at implementing an end-to-end adaptive AI-based solution for patients suffering from hearing loss, with a focus on the embedding the developed algorithms. WHO: https://www.who.int/deafness/estimates/en/ Hear-it, 2019: https://www.hear-it.org/untreated-hearing-loss-eu-costs-more-whole-eu-budget de Kervasdoué, J., & Hartmann, L. (2016). Impact Economique du Déficit Auditif en France et dans les Pays Développés. UNSAF

Early, easy and rapid diagnosis of coronavirus disease 2019 (COVID-19) is of the utmost importance but remains challenging. Breath analysis is an innovative, non-invasive, real-time, point-of-care technique for detecting volatile organic compounds (VOCs) in expired breath with potential for use in diagnosis and large-scale screening. Our consortium has previously shown in a study with mass spectrometry breath analysis that patients with severe COVID-19 have a discriminating “breathprint”, which includes at least a set of four putatively identified VOCs. Electronic noses (eNoses) are portable devices for breath analysis consisting of sensor array and pattern recognition algorithm to generate signal patterns that are already used in clinical research. Our project is aimed at (i) providing a formal identification of the COVID-19 specific VOCs, (ii) performing an extensive investigation of the performance of the different eNoses and sensors for the detection of these VOCs to select the best sensors for COVID-19 diagnosis, using laboratory investigations and data from ongoing clinical trials, (iii) optimizing the eNaiR software for the VOC signature detection and the COVID-19 status prediction, and (iv) designing and setting-up validation clinical trials in independent patient cohorts (observational clinical trials) with the optimized sensors and analytical strategy. This project brings together a consortium of experts in analytical sciences (mass spectrometry and eNoses), data processing, clinical research in patients with severe infections and clinical research in exhaled breath analysis.

The MOBILEX Challenge tackle autonomous navigation of vehicles in complex environments. To meet the requirements of the Mobilex challenge, CEA and Magellium have combined their competencies in the fields of perception, localization and navigation in complex environments. For over 20 years, Magellium has been supporting the CNES in the development of perception and localization components in order to increase the autonomy of rovers; some of them are embedded in major missions (ExoMars, MMX, ...). This expertise, accredited by the CNES, allows us today to support large transport (Renault, SNCF, etc.) and defence companies (Arquus, Nexter, DGA, etc.) in the use and the deployment of these technologies. The Interactive Robotics Department (SRI) of the CEA-LIST brings its renowned excellence in the fields of command and control, autonomous navigation and robotics system, developed during numerous French, European, industrial or in-house projects, whether in the nuclear, transite logistics or agricultural fields. We thus propose a consortium mastering all the technologies necessary for the success of the project, with a strong competence in robotics and a high scientific level, which will be able to offer solutions to the problems of increasingly complex challenges and will participate in the emergence and industrialization of state-of-the-art algorithms. To successfully complete the challenge, we have defined 5 objectives: (1) To equip the platform with perception and localization capabilities, (2) To integrate an innovative navigation function adapted to the characteristics of the challenge, (3) To provide the robot with a capability to characterize the traversability of the terrain, (4) To develop a mobile robotic platform capable of operating in complete safety, (5) To communicate, disseminate and promote the results of the project. The methodology proposed will be based on the adaptation of existing components, the increase in maturity and the robustification of innovative functions. Challenge #1, which will require the equipment and the handling of the platform, will mainly consist in adapting and integrating existing functionalities brought by each partner to ensure the autonomous mobility of the platform in an environment with limited complexity. Meeting challenge #2 will however require the addition of complementary perception modalities and the integration of innovative algorithms developed specifically. A redundancy will be ensured by the robustification of the existing functionnalities. Finally, in order to face the challenge #3, we will work mainly on the robustification and the improvement of the performances of the whole system set up during the first two challenges. Thus, the envisaged tasks allow to achieve a rise in maturity throughout the project of the proposed hardware and software system, while minimizing the risks and in the respect of the phasing of the project marked by the three challenges. Beyond the technical challenge addressed by innovative technologies and a high-standing consortium, a particular effort will be made on communication, dissemination and valorization, key elements of our proposal. The partnership composition of our project allows us to reach a large scientific, industrial and operational audience, through adapted communication strategies and through dedicated events (exhibitions, seminars, conferences, thematic days, etc.). Finally, scientific and technical valorization activities will be carried out.