EN FR
EN FR




Section: Partnerships and Cooperations

National Initiatives

ANR Vulcain (06/2008-06/2011)

Participants : Marie Durand, François Faure.

We participate to the ANR Vulcain project (http://vulcain.ujf-grenoble.fr/ ), which purpose is to evaluate the vulnerability of buildings such as industrial facilities undergoing explosions of projectile impacts. Marie Durand has implemented discrete element models in GPU in order to speed up concrete fracturing simulations, and an article has been submitted to the European Journal of Environmental and Civil Engineering.

ANR RepDyn (01/2010-12/2012)

Participants : Marie Durand, François Faure.

We participate to the ANR RepDyn project, startied at the beginning of 2010, in collaboration with CEA, EDF, Laboratoire de Mécanique des Structures Industrielles Durables (LaMSID), and ONERA. The purpose of this project is to enhance the performance of discrete elements and fluid computations, for the simulation of cracks in nuclear reactors or planes. Our task is to propose GPU implementations of particle models, as well as load balancing strategies in the context of multi-core, multi-GPU hardware. Marie Durand is doing a PhD thesis on this task.

ANR ROMMA (01/2010-12/2013)

Participants : François Faure, Stefanie Hahmann.

The ANR project ROMMA has been accepted in 2009 and started in January 2010. The partners of this project are academic and industry experts in mechanical engineering, numerical simulation, geometric modeling and computer graphics. There are three academic members in the consortium: the LMT in Cachan, G-SCOP and LJK (EVASION and MGMI teams) in Grenoble. There are four industrial members: EADS, which coordinates the project, SAMTECH, DISTENE and ANTECIM. The aim of the project is to efficiently and robustly model very complex mechanical assemblies. We are working on the interactive computation of contacts between mechanical parts using GPU techniques. We will also investigate the Visualization of data with uncertainty, applied in the context of the project.

ANR SOHUSIM (01/2010-12/2013)

Participants : Ali Hamadi Dicko, François Faure.

Sohusim (Soft Human Simulation) is done in collaboration between: EVASION (INRIA), Fatronik France (TECNALIA), DEMAR (INRIA), HPC PROJECT and the CHU de Montpellier.

This project deals with the problem of modeling and simulation of soft interactions between humans and objects. At the moment, there is no software capable of modeling the physical behavior of human soft tissues (muscles, fat, skin) in mechanical interaction with the environment. The existing software such as LifeMod or OpenSim, models muscles as links of variable length and applying a force to an articulated stiff skeleton. The management of soft tissues is not taken into account and does not constitute the main objective of this software.

A first axis of this project aims at the simple modeling and simulation of a passive human manipulated by a mecatronics device with for objective the study and the systems design of patient's manipulation with very low mobility (clinic bed). The second axis concentrates on the detailed modeling and the simulation of the interaction of an active lower limb with objects like orthesis, exoskeleton, clothes or shoes. The objective being there also to obtain a tool for design of devices in permanent contact with the human who allows determining the adequate ergonomics in terms of forms, location, materials, according to the aimed use.

Dicko Ali Hamadi is a Ph.D. student within EVASION team. His works turns around the problems in SOHUSIM project. He is co-tutored by Olivier Palombi in EVASION, in collaboration with Benjamin Gilles in DEMAR.

PlantScan3D (ARC INRIA 09/2009-09/2011)

Participants : Dobrina Boltcheva, Marie-Paule Cani, Cédric Zanni.

This project is in collaboration with Vitual Plants and Galaad teams. Its objective is to develop the use of laser scanner for plant geometry reconstruction, in partnership with biologists-agronomists from several teams in France and Europe. Our last contributions include the development of new representation for the plant enabling to use the skeleton and thickness information computed by the other teams from scanner data.