Inria Project Labs, major interdisciplinary research programmes
Inria Project Labs' initiatives enable the launch of ambitious research projects directly linked with the institute. These programmes are often interdisciplinary and call on a wide range of skills. Aim: mobilising and highlighting the expertise of Inria researchers around key challenges.
Inria Project Labs (IPL) operate as inter-team project teams. They are therefore peer-reviewed—in general by foreign experts—and launched for a four-year period. Led by a science officer, Inria Project Labs' initiatives are formalised around a specific research issue with a predefined programme and objectives. They call on the combined "brain power" of all researchers in Inria's various project teams, as well as the intellectual resources of academic or industrial partners.
IPL initiatives allow researchers to organise their work around major issues the Institute wants to concentrate on. A small number of initiatives are launched at a time and benefit from considerable resources. Proposals are examined by the Research Department. Initiatives are reviewed at the halfway mark. A second review, which presents the results obtained, takes place at the end of the project.
BCI-LIFT: a new generation of Brain-computer interfaces
Brain-computer interfaces (BCI) are human-machine interaction systems that rely on the analysis of brain activity. They lead to new paradigms for interaction, communication, diagnosis and therapy. BCI-LIFT (Learning, Interaction, Feedback, Training) aims at a new generation of BCI that are simple to handle, efficient, and usable by a large number of people. Our consortium involves experts in signal processing, statistical learning, human-machine interaction, virtual and augmented reality, human motor control, electrophysiology, cognitive neurosciences and brain-computer interfaces. With usability as driving objective, BCI-LIFT will develop new user-centered approaches, notably for training and for proficiency evaluation.
Targeted applications include:
- in the clinical domain, a neurorehabilitation training platform for upper-limb motor control restoration,
- for the general public, an interactive brain-activity visualization tool
- in view of operating standard applications through BCI, a GUI-level interaction protocol.
CAPPRIS: Protection of Privacy Rights in the Information Society
Cappris is an Inria Project Lab initiated in 2013. The general goal of Cappris is to foster the collaboration between research groups involved in privacy in France and the interaction between the computer science, law and social sciences communities in this area. In order to reach its goals, Cappris will carry out two kinds of actions :
- Joint Research Actions to investigate specific research topics following a collaborative and interdisciplinary approach. Three Joint Research Actions have been launched: the first one is dedicated to the notion of consent, the second one aims to devise a privacy reference architecture and the third one focuses on privacy assessment.
- Networking actions to favour the emergence of a research community on privacy and enhance the interest of researchers and the public in this fast evolving domain.
The outputs of the first line of actions are research results whereas the networking actions will take the form of joint events (meetings, visits, workshops, etc.) as well as wider audience publications and events.
Even if its goal is to provide general techniques with a potentially broad impact, Cappris will consider different contexts and concrete case studies to ensure the relevance and significance of its results. To reach this aim, three classes of case studies have been selected: Online Social Networks (OSN), Location Based Services (LBS) and Electronic Health Record Systems (EHR), which correspond to application domains with great impact on society.
Project teams : PRIVATICS
- Leader : Daniel le Metayer
C2S@Exa: Computer and Computation Sciences @ exascale
It is nowadays recognized that a multidisciplinary approach is required to overcome the challenges raised by the development of highly scalable numerical simulation software that can exploit computing platforms offering several hundreds of thousands of cores. To achieve this goal, the C2S@Exa initiative gathers
- computer scientists that study programming models, and develop environments and tools for harnessing massively parallel systems,
- algorithmicists that devise numerical kernels and core solvers, and develop generic libraries in order to take benefit from all the parallelism levels with the main goal of optimal scaling on very large numbers of computing entities
- numerical mathematicians that are studying numerical schemes and develop parallel solvers for systems of partial differential equations in view of the simulation of complex physical problems.
- Leader : Stéphane Lanteri + 33 4 92 38 77 34
The current rate of fossil fuel usage and its serious adverse environmental impacts (pollution, greenhouse gas emissions, ...) leads to an energy crisis accompanied by potentially disastrous global climate changes. The research of alternative energy sources is thus of crucial importance. Controlled fusion is one of the most promising alternatives to the use of fossil resources, potentially with a unlimited source of fuel.
Controlled nuclear fusion can be considered as an example of grand challenge in many fields of computational sciences from physical modeling, mathematical and numerical analysis to algorithmic and software development and several Inria teams and their partners are developing mathematical and numerical tools in these areas.
The goal of FRATRES is to organize these developments on a collaborative basis in order to overcome the current limitations of today numerical methodologies. The ambition is to prepare the next generation of numerical modeling methodologies able to use in an optimal way the processing capabilities of modern massively parallel architectures.
This objective requires close collaboration between:
- applied mathematicians and physicists that develop and study mathematical models of PDE;
- numerical analysts developing approximation schemes;
- specialists of algorithmic proposing solvers and libraries using the many levels of parallelism offered by the modern architecture and computer scientists.
The project will contribute in close connection with National and European initiatives devoted to nuclear Fusion to the improvement and design of numerical simulation technologies applied to plasma physics and in particular to the ITER project for magnetic confinement fusion.
Project-teams : CASTOR, KALIFFE, IPSO, TONUS
Partners : IRFM-CEA, Max Planck Institute-IPP Garching, LJLL-Jussieu, IMT-Toulouse
- Leader : Hervé Guillard, +33 4 92 38 77 96
MORPHOGENETICS: Deciphering morphogenesis at multiple scales
Morphogenetics is aimed at understanding how shape and architecture in plants are controlled by genes during development. To do so, we will study the spatio-temporal relationship between genetic regulation and plant shape utilizing recently developed imaging techniques together with molecular genetics and computational modeling. Rather than concentrating on the molecular networks, the project will study plant development across scales. In this context we will focus on the Arabidopsis flower, currently one of the best-characterised plant systems.
Partners : UMR RDP (ENS-Lyon, Inra, CNRS - Lyon) and RFD (UMR Physiologie Cellulaire Végétale CEA-INRA-CNRS - Grenoble)
- Leader : Christophe Godin - Tél. : 04 67 61 65 77
MULTICORE: a novel approach based on virtualization and dynamicity
Multicore processors are becoming the norm in most computing systems. However supporting them in an efficient way is still a scientific challenge. This large-scale initiative introduces a novel approach based on virtualization and dynamicity, in order to mask hardware heterogeneity, and to let performance scale with the number and nature of cores. It aims to build collaborative virtualization mechanisms that achieve essential tasks related to parallel execution and data management. We want to unify the analysis and transformation processes of programs and accompanying data into one unique virtual machine. We hope delivering a solution for compute-intensive applications running on general-purpose standard computers.
- Leader: Gilles Muller - Tel.: + 33 1 44 27 88 52
HEMERA: developing large scale parallel and distributed experiments
Grid'5000, launched in 2004 by Inria, has become an indispensible tool in software experimentation. Grid'5000 represents nearly 3,000 processors in more than 1,000 nodes, located on 9 French sites, with extensions in the Netherlands, Japan, Luxembourg and Brazil. It is the largest shared computer network in Europe, reserved for research in computing. The aim of this initiative is to experiment with large-scale research problems, such as grid optimisation algorithms, studying the robustness of peer-to-peer networks or simulations in the fields of hydrogeology or energy. This means leading and extending the scientific community around Grid'5000.
Project teams: ALGORILLE, ASAP, ASCOLA, ASTRE, CEPAGE, DOLPHIN, GRAAL, GRAND-LARGE, KERDATA, MESCAL, MYRIADS, OASIS, REGAL, RESO, RUNTIME, SAGE, in partnership with the LSIIT (Strasbourg), the LAAS and IRIT (Toulouse).
- Head of the initiative: Christian Perez - Tel.:+ 33 4 72 72 84 34
PAL: An initiative dedicated to assisting people
This goal is to create an infrastructure that can enable numerous Inria teams to work on technologies for assisting people, to experiment them with users and learn from their experience. The goal is to offer the elderly or disabled more autonomy and a better quality of life. Four research topics will be studied: estimating the degree of frailty of elderly people using non-invasive sensors in order to prevent falls or detect signs of malnutrition; developing mobility equipment such as walkers, wheelchairs; assisting people in getting up, for example from a bed; studying the best means of communication for preserving social links.
REGATE: modelling the reproductive function (IPL completed)
Reproduction and its biological mechanisms, in humans and animals, are of increasing interest for researchers. This major physiological function is complex and tightly controlled. It involves the hypothalamus (in the central nervous system), the pituitary gland (an endocrine gland) and the gonads (testicles and ovaries). Researchers are studying the modelling, simulation and control of these different organic levels (from cells to tissue) and their interactions. Multiple temporal scales are concerned: from hormonal communications lasting a few minutes to the 28-day menstrual cycle, from puberty to menopause. Applied mathematics, IT, control theory, but also physiology, cellular and molecular biology are all fields involved in the project.
Project teams: SISYPHE, CONTRAINTES and the following partners: Jacques-Louis Lions laboratory (University Pierre & Marie Curie), BIOS (Biology and Bioinformatics of Signalisation systems) and BINGO (Biology INtéGrative de l'Ovaries/Integrative Biology of the Ovary) teams from the INRA research centre in Tours and the Unit of Theoretical Chronobiology of the ULB (Université Libre de Bruxelles).
- Head of the large-scale initiative: Frédérique Clément - Tel.: +33 1 39 63 53 83
SOFA-INTERMEDS: simulating surgical procedures (IPL completed)
Surgical simulators are being increasingly used to practice complex procedures or learn new techniques in highly realistic conditions. Researchers are developing algorithms dedicated to medical simulation and prototyping of medical simulators. These algorithms are integrated in the SOFA software platform, which can also be used to compare and validate results. A series of medical simulators should be developed, for example, to simulate minimally invasive surgery in ophthalmology, interventional radiology or cardiology.
Project teams: ALCOVE, ASCLEPIOS, EVASION, BUNRAKU, MAGRIT, MOAIS and VISAGES and the following clinical partners: CHR de Lille and CHR de Nancy hospitals, IRCAD and the Massachusetts General Hospital in Boston.
More realistic medical simulators
Sofa InterMeds is a large-scale initiative that aims to develop algorithms dedicated to medical simulation and prototyping of simulators. " This means adding new functions to the SOFA simulation platform, in order to improve its performance . " underlines Stéphane Cotin, Head of Sofa InterMeds and the S.H.A.M.A.N. project team, Inria Lille - Nord Europe " In a nutshell, we want to upgrade the software through our research. Recently we have developed a highly realistic prototype simulator for learning techniques in ophthalmology with the CHR de Lille hospital. We modelled tissue behaviour in order to calculate its deformation in real time and provide a very precise image for the manipulator ".
With the participation of six project teams, Inria's contribution to this research project is considerable " emphasises Stéphane Cotin. Another project goes beyond learning and is being conducted in partnership with clinicians of the CHU de Nancy hospital who are specialised in interventional radiology. This involves developing a tool to ensure better planning of treatments for strokes. To do this, researchers have designed a system that integrates all the patient's imaging data (MRI, scans). Furthermore, in partnership with the IRCAD (Research Institute against Digestive Cancer) in Strasbourg, the team is attempting to develop an augmented reality system that can be used to visualise the internal structures of the liver (vascular network, tumours...) during an operation (hepatectomy)."
SYNCHRONICS: improving the programming of embedded systems (IPL completed)
Initiated in January 2008, Synchronics aims to design a new programming language in order to develop critical embedded systems. These electronic and computer systems are used in avionics, nuclear power, automobiles, etc. This language will be synchronous, as is generally the case in programming embedded systems, and will use specific semantics, like the languages usually used to model them. The aim is also to define a compilation method for modern architectures, mainly multicore implementations.
Project teams: LIENS, PROVAL, ALCHEMY, POP ART, S4, Vérimag laboratory (Synchrone team)
CardioSense3D: a digital heart adapted to each patient (IPL completed)
Cardiosense3D, the first large-scale initiative launched by Inria, ended in 2009. This original interdisciplinary research project will be prolonged via a European project on the digital heart called EuHeart. The goal: to simulate the workings of the human heart in order to understand certain heart diseases and improve prevention, diagnosis and patient therapies. The academic partners, clinicians and medical industries involved have designed a heart simulator. It is based on physiological principles at the cellular, tissue and organ levels. The parameters of each model are set for an individual patient, a necessary approach for clinical use. The simulator is being tested with animal data and on patients who suffer from arrhythmia.
Project teams: ASCLEPIOS, MACS, SISYPHE, REO and international partners, notably the clinical teams at Guy's Hospital (London) and NIH (Washington).
- Head of the large-scale initiative: Hervé Delingette - Tel.: +33 4 92 38 77 64
COLAGE: controlling the growth of bacteria (IPL completed)
This large-scale initiative combines approaches in systems biology and synthetic biology. It involves studying and controlling the growth of bacteria. Innovative IT tools and quantitative experimental studies are used to explore the variability of bacterial growth in different environments. It is hoped that understanding and controlling the growth and aging of bacteria will lead to applications in biotechnologies and the creation of new drugs.
Project teams: ALCHEMY, COMORE, CONTRAINTES, IBIS and the following partners: experimental cellular biology teams from the INSERM (Université Paris Descartes) and the CNRS (LAPM, Université J. Fourier Grenoble).
- Head of the large-scale initiative: Hugues Berry - Tel.: + 33 4 72 43 72 84
FUSION: simulating nuclear fusion (IPL completed)
Thermonuclear fusion is a promising field for electricity production in the long term. Construction of an experimental reactor in Cadarache (France) is currently being studied, in the context of the international ITER project. Fusion requires temperatures of one million degrees Celsius and magnetic confinement of plasma (ionised gas). In 2006 and 2007, several Inria teams developed tools to analyse the behaviour of plasma within the framework of the collaborative research project Fusion. This large-scale initiative builds on these results in order to achieve two objectives: continuing the development of nuclear fusion simulation tools in general, and of ITER in particular, and building a community of mathematicians and computer scientists around this specific area of physics.
Project teams: CALVI, METALAU, BACCHUS, HIEPACS, SIMPAF, SMASH, PUMAS, GAMMA, APICS, TROPICS, university teams and CNRS centres at Nice, Toulouse and Paris 6, and the CEA of Cadarache.
- Head of the large-scale initiative: Eric Sonnendrücker
These articles could interest you:
To find out more
Daniel Le Metayer