Life is tough for planktonic copepods, constantly washed by turbulent flows. Yet, these millimetric crustaceans dominate the oceans in numbers. What have made them so successful? Copepod antennae are covered with hydrodynamic and chemical sensing hairs that allow copepods to detect preys, predators and mates, although they are blind. How do copepods process this sensing information? How do they extract a meaningful signal from turbulence noise? Today, we do not know. C0PEP0D hypothesises that reinforcement learning tools can decipher how copepod process hydrodynamic and chemical sensing. Copepods face a problem similar to speech recognition or object detection, two common applications of reinforcement learning. However, copepods only have 1000 neurons, much less than in most artificial neural networks. To approach the simple brain of copepods, we will use Darwinian evolution together with reinforcement learning, with the goal of finding minimal neural networks able to learn. If we are to build a learning virtual copepod, challenging problems are ahead: we need fast methods to simulate turbulence and animal-flow interactions, new models of hydrodynamic signalling at finite Reynolds number, innovative reinforcement learning algorithms that embrace evolution and experiments with real copepods in turbulence. With these theoretical, numerical and experimental tools, we will address three questions: Q1: Mating. How do male copepods follow the pheromone trail left by females? Q2: Finding. How do copepods use hydrodynamic signals to ‘see’? Q3: Feeding. What are the best feeding strategies in turbulent flow? C0PEP0D will decipher how copepods process sensing information, but not only that. Because evolution is explicitly considered, it will offer a new perspective on marine ecology and evolution that could inspire artificial sensors. The evolutionary approach of reinforcement learning also offers a promising tool to tackle complex problems in biology and engineering.

The goal of the SINUMER project is to build a numerical simulator to capture the key physical mechanisms of mucus transport in human lungs, and use it to progress on the understanding of severe asthma and Chronic Obstructive Pulmonary Diseases (COPD). These chronic respiratory diseases affect nowadays hundreds of millions of people, and there is still no curative treatment available. Moroever this worldwide burden is constantly growing because of the increase of external toxic agents and pollutants in the air of modern urban societies. At the roots of these bronchial disorders, lies the bronchial epithelium, where is transported mucus, a complex fluid powered by the coordinated beating of billions of microscopic cilia carried by the epithelial cells. These cilia can synchronize their beating and produce typical waves which transport the mucus through the bronchial tree. This transport is determined by largely unknown physical mechanisms involving a hydrodynamic coupling between the ciliary beats and the surrounding multiphase non-Newtonian fluid. The numerical environment built in SINUMER will be the first of a kind in respiratory research at this level of realism, taking advatange of an interdisciplinary approach which combines complementary experiences and know-how on mucociliary transport, independently acquired in each discipline, and a synergistic approach linking computational fluid dynamics, biophysical experiments and medicine. The numerical environment will be used towards two objectives: -the first one is to better understand the multi-scale biophysical mechanisms involved in the transport of mucus, from the individual cilia beating to their macroscopic collective motion, by unraveling macroscopic laws to link the microscopic scale with the clinical observables. This will allow to elucidate the role of the epithelium parameters: density, frequency, direction, beating patterns and coordination of cilia, and mucus rheology; - the second one is to provide a numerical diagnosis tool for chronic respiratory diseases, based on the identification of relevant biophysical markers of mucociliary alteration. The prospects are therefore to study in-vitro and in-silico innovative strategies for the development of therapeutic curative and not only symptomatic. The perspective is thus to enrich the knowledge on chronic airway diseases, their phenotypes and their diagnoses thanks to the new light brought by computational fluid dynamics, and to provide a numerical tool, reliable and efficient for prognostic and therapeutic diagnostic steps in human clinical practice. Such an integrated in-silico approach is strongly expected to yield innovative strategies for curative care of chronic respiratory diseases. Moreover, the longer-term ambition of the SINUMER project is to go towards a patient-personalized physical description of the pathologies which relies on the individualized numerical ‘clearance footprint’ of each patient to progress on the major issue of ‘personalized medicine’. An important outcome of SINUMER for the medical community is to provide in-depth modifications of how chronic respiratory diseases are clinically considered, by resetting the focus on the mucociliary transport.

The Chair project deals with the development of efficient Latice-Boltzmann Methods (LBM) for the simulation of flows in realistic industrial applications in the field of aerodynamics, aeroacoustics and heat transfer, e.g. installed aircraft engine, car engine under hood, full car/aircraft with moving elements. This Chair is supported by three leading companies, namely Airbus, Renault and Safran. The Chaire candidate is Prof. Pierre Sagaut from M2P2 Laboratory (Aix-Marseille Université/CNRS/Ecole Centrale Marseille). The research program is organized in four axes, each one aiming at removing a scientific lock of major importance for the use of LBM for applications. These methods have become essential in the field of Computational Fluid Dynamics because of their very high efficiency compared to classical tools based on the resolution of Navier-Stokes equations, in particular for low-Mach massively separated flows. Their extension to highly compressible flows and the possibility of treating increasingly complex cases in terms of geometry (including deformable solid objects and/or with arbitrary kinematics) is therefore a crucial issue in terms of deployment strategy of these methods for engineering and raises many theoretical questions on numerical methods and physical models. The four research themes that structure the Chair are: - numerical modelling of solid walls, including heat transfers - the development of efficient LBM methods for the simulation of highly compressible flows and shock capturing - development of LBM methods for flow simulation in the presence of deformable solids and/or in arbitrary motion - modelling turbulence within the framework of LBM methods, taking into account their specificities (nested uniform Cartesian meshes, discretization errors, etc.) The governance of the project is organised on three levels: - a Technical Committee, which ensures the operational follow-up on a monthly basis of all the tasks defined within the themes. This committee is composed of the chairholder, the heads of each research theme, and a representative of each industrial partner. - the Steering Committee, which meets twice a year and discusses and approves the main orientations and monitors the Chair's budget. - an International Advisory committee, composed of leading French and foreign researchers on the project's research themes, which will meet at least three times over the four years of the project. Its role will be to assist the steering committee in making decisions by providing an analysis of the international context and a prospective vision of LBM methods, and also by recommending international collaborations. In addition to the work of the permanent researchers at M2P2, the resources implemented consist of 4 thesis funding (one on each theme) and 16 years of post-doctorate (spread over the four themes). Making research results available to a wide public is a strong focus of the project. In addition to the fact that all articles will be published in "Gold open access" format, the data and results will be published on a website dedicated to the Chair. This site will also contain an intranet to allow the exchange of non-public data between project members. This mechanism will be reinforced by the organization of two international workshops, to which the main international research groups involved in the four research themes will be invited. These events will be supplemented by doctoral training organized in conjunction with the Doctoral College at the Aix-Marseille site.

During embryonic development, cells must perform large-scale coordinated motion to shape functional tissues and organs. The development of computational models for such tissue flows may radically change the way we understand and control tissue morphogenesis. However, we still lack fundamental insights into the underlying mechanics of these processes. At the tissue level, these flows result from a complex interplay between growth, active stresses and complex mechanical properties. The previously overlooked role of mechanical properties, also referred to as rheology, has recently received growing attention as rheology has been shown to play a key role in various morphogenetic events. Yet, accurate measurements of tissue rheology in conditions relevant to morphogenesis remain currently challenging. Moreover, the general role of tissue rheology and its coupling to growth and active stresses during morphogenesis remains unclear. In FluidEmbryo, we aim to leverage recent advances in machine learning and computational fluid dynamics (CFD) to determine how complex mechanical properties of embryonic tissues can drive tissue flows shaping organs, with two objectives: - Objective 1 aims to build effective mechanical models for embryonic tissues. We will infer rheological properties and active stresses in embryonic tissues from microfluidic experimental data using physics-informed neural networks (PINNs). - Objective 2 aims to determine how mechanical properties can sculpt tissues. We will use high performance CFD simulations coupling complex rheology, growth and active stresses, to (i) identify morphogenetic processes that can be driven by tissue rheology in generic configurations and (ii) determine the role of tissue rheology during the axis formation of embryonic organoids. These two objectives will be pursued in close collaboration with two experimental groups with internationally recognized expertise in tissue morphogenesis and embryonic organoids. As a continuation of this project, our mechanical models will later be coupled with data-driven biochemical models, opening the way to multi-physics computational simulations for tissue engineering.

When a weakly diffusive chemical substance is transported through a turbulent or a porous medium, the resulting concentration field is characterised by a structure made of filaments stretched and folded by the flow. Many organisms have to navigate this complex landscape to find their way to the emitting source, from the bacteria living in our soils, to the cells migrating in our tissues, to the insects pollinating our gardens. These organisms interact with each other and often move together in groups, which allows them to reach the source more efficiently than if they were alone. This project aims at deciphering the mechanisms of this cooperation. To this end, we will combine computational models of collective motion, direct numerical simulations of scalar transport in turbulent or porous media flows, and optimization techniques from artificial intelligence. Virtual agents will interact within a simulated environment that reproduces the key features of chemical signals found in the wild. To discover the best cooperation strategies, these agents will be trained at navigation using deep reinforcement learning. Equipped with this unique numerical platform, we will perform in silico experiments to answer three questions: - Q1: Which individual behaviours allow global navigational abilities to emerge? - Q2: How does scalar transport in disordered flows constrain navigation and communication strategies? - Q3: Which features of collective navigation can be conserved across systems and scales? This mechanistic approach to living systems will uncover the fundamental principles of collective navigation and will contribute to the emerging field of smart active matter. It will also provide guidelines for the design of robotic swarms able to navigate autonomously.