The main objective of COMBY project is to develop fibered Fabry-Pérot (FP) resonators, with GHz free spectral range and versatile for the generation of optical frequency combs (OFCs) and microwave signals. This multidisciplinary project, at the interface between materials, non-linear photonics and microwaves, aims for several scientific and technological breakthroughs by exploring applied and exploratory research directions. The COMBY project is structured on 5 major scientific objectives: (1) Develop a mature and versatile FP resonator manufacturing platform able of achieving record quality factors (Q>10^9). (2) Gain a better understanding of the emergence of Kerr frequency combs in FP resonators using theoretical and numerical tools (especially for multimode configurations). A thorough understanding of the physical processes involved in the emergence of frequency combs is indeed essential for the success of the project. (3) Generate spatially multiplexed and mutually coherent OFCs within the same FP resonator. (4) Develop a compact laser demonstrator locked on an FP cavity and having an integrated linewidth in the range of 30 Hz. (5) Create a microwave or millimeter wave frequency generator based on synchronized combs, with phase noise below -120 dBc/Hz at 1 kHz of a 10 GHz carrier. The COMBY project will boost the emerging synergy around fiber FP resonators at the national level and gain international recognition in this highly promising field, as evidenced by the numerous recent publications on the subject. Furthermore, our consortium's holistic approach, encompassing fabrication, modeling, and experimental demonstrations, will provide us with all the tools to raise the technological maturity level (TRL) of these devices.

Parkinson’s disease is a multisystem progressive neurodegenerative disorder with pathological hallmarks, including dopaminergic neuronal degeneration in the nigrostriatal pathway, abnormal iron build-up and alpha synuclein protein accumulation forming Lewy body inclusions. Over the last years, neuroimaging methods have provided invaluable insights into some mechanisms as iron load and the neurodegeneration. However, still, no imaging method has yet been able to inform about the a synuclein aggregation and spreading. The aim of the ImSynPark project is to develop a chemical exchange saturation transfer magnetic resonance imaging (CEST) method for the in vivo quantification of brain alpha synuclein protein with clinically compatible acquisition times. CEST imaging, a part of what is known as metabolic MR imaging, can provide, with a good spatial resolution, an indirect in vivo quantification of proteins implied in several neurodegenerative disorders The originality of our approach consists in targeting new pathophysiological processes of PD by acquiring quantitative MRI data not currently available in national or international databases, and in developing complete protocols for processing these data. The project is built as a translational project starting by in vitro experiments right up to the first clinical tests using 3T and 7T MR machines. The project includes 3 main steps: in vitro experiments and calibrations, in vivo investigations using animal models and clinical tests. The project takes advantage of a unique pool of expertise: 3 research teams and 2 imaging platforms, both members of the France Life Imaging (FLI) network. The project merges three complementary topics: neurobiology, biophysics and image processing. All partners have the required experience to follow the roadmap and to respond to challenges of this project and produce a major breakthrough.

STEFI aims to identify the innovative potential of start-ups in order to better manage the seed and funding support, especially public funding. Thanks to the co-construction of data bases, the research team and HDFID assess the value relevance of early stage signals of innovation and how they interact with the level of funding provided by backers. The research aims to explore what can explain the presence of a variety of funders in early stage equity funding and whether public funds play a pivotal role in this process. A better understanding of start-up trajectories will help to anticipate the performance in terms of growth, amount raised, and survival. Transfer competencies between the research team and the company will be based on the co-construction of data bases and new innovation indicators. The meaning given to these indicators will also be a joint work. Moreover, researchers usually work on a single financial vehicle. The new data bases will provide the combination of a large range of investors, allowing to improve theoretical knowledge on innovation signals and funding trajectories of start-ups.

Biodiversity loss in conventional farmland is one of the most pressing issues that humanity has to face. Using the approach of living labs that promote the involvement of citizens in science, this project strives to collectively develop field-to landscape management, mainly by floral enrichment, and bioindicators about the conservation state of farmland biodiversity. In this project, we will focus on cereal fields along a climatic gradient from the mild Atlantic climate (western France) to the more continental climate of central Europe (western Czech Republic). In line with the concept of ecological intensification, conservation of biodiversity aims at maximizing ecosystem services, here pest and weed controls and pollination, and to minimize disservices (presence of weeds and pests, loss of crop yields). The European climatic scale investigated will help to provide European-wide solutions for adaptation to land-use and climate changes. Along a climatic gradient, it is expected that the climatic context plays a major role on the potential of ecosystem services in each area. Therefore, to be able to effectively design plant floral enrichment that supports pest and weed control as well as pollination at the European scale, a study is needed on a large spatial and climatic gradient that would include a large range of taxa (wild flowers, slugs, aphids, parasitoids, hyperparasitoids, spiders, rove beetles, carabids, dung beetles, syrphids, butterflies, and bees) and landscape contexts. Simultaneously managing multiple ecosystem services requires understanding the mechanisms underlying ecosystem service interactions. The approach we propose to tackle this problem is multidisciplinary and based on the combination of the living lab concept, citizen-based means of field data collection called BioBlitz, and manipulated field experiments by floral enrichment that will reflect the results of the living lab, BioBlitz and scientific data. Among the outcomes of the project, living labs will be established in the four countries involved (France, Belgium, Germany and Czech Republic). To assess biodiversity, we will develop two kinds of multi-taxon-based integrated indicators. Finally, scenarios of adding diversity within, nearby and in the surroundings of the fields in order to optimize diversity in agro-ecosystems at the farm/landscape scales will be co-developed in living labs with farmers to engage them, in protecting biodiversity and ecosystem health. Our approach will contribute to the knowledge needs specified in the Themes 1 and 2 of the Call document Biodiversa+. Our project will provide tools adapted to different climatic/and local to landscape practices as the chosen countries are contrasted in their climate, landscape history and agricultural practices. The consortium includes 5 academic partners, a company and 2 stakeholders, that will ensure the dissemination of the results to the farmers.

The growing utilization of artificial intelligence (AI), like deep neural networks, in embedded real-time systems poses many challenges. One of these is the very large quantity of data processed by these algorithms, which impacts their execution time and power consumption on resource-limited platforms. To address this problem, software and hardware approaches are increasingly being adopted. On the one hand, hardware accelerators (e.g., FPGA, GPU, TPU) make it possible to optimize matrix calculation and process the calculation divided into several subtasks in parallel. On the other hand, software techniques (e.g., network pruning, and quantization) make it possible to reduce the size of the network. This project seeks to explore the potential of hardware accelerators and software optimization techniques in achieving efficient and accurate AI inference in embedded real-time systems. Our approach involves developing new scheduling policies specifically designed for hardware accelerators, which can dynamically regulate the balance between time, energy, and accuracy. These policies require an offline optimization framework, which, in the iterative process, can determine the level of compression applied to the neural networks to ensure the schedulability of the resulting neural net tasks and their energy and accuracy indicators. We plan to evaluate the proposed framework on an autonomous F1TENTH car and mini rover with an Edge TPU accelerator.