Simulation Open Framework Architecture
||A major contributor to this article appears to have a close connection with its subject. (January 2017) (Learn how and when to remove this template message)|
|Developer(s)||Inria, CNRS, USTL, UJF, MGH|
|Initial release||February 2007|
16.08 / August 2016
|Written in||C++, Python|
|Operating system||Microsoft Windows, OS X, Linux|
|Type||Physics engine, Medical simulation, Framework|
Simulation Open Framework Architecture (SOFA) is an open source framework primarily targeted at real-time physical simulation, with an emphasis on medical simulation.
It is mostly intended for the research community to help develop newer algorithms, but can also be used as an efficient prototyping tool or as a physics engine.
The idea of creating an open-source platform for physics-based simulation first arose in the rooms of the CIMIT in Boston in 2000. It took about four years before a white paper describing the concept of the architecture of SOFA was written in 2004. For the first time, this white paper foreseen a collaborative project between the CIMIT and Inria (Alcove team in Lille). Soon, further Inria teams joined the project in Sophia-Antipolis and Grenoble. Early in 2005, the first live demonstration of SOFA was performed at the occasion of MMVR in Los Angeles. Subsequently, the American Department of Defense decided to fund the development of SOFA while Inria actively supported the project with the first SOFA engineers.
The SOFA project really started in 2006 in Boston.
Then, the implementation of the SOFA framework intensively started with new engineers, the first PhD students working on SOFA (CSIRO, Inria). In 2007, a publication describing in details the structure and the concepts of SOFA was accepted at MMVR. Since then, many engineers and PhD students contributed and experienced SOFA. Many of them are still in the community!
Today, SOFA gathers more than 10 years of research in physics simulation. Many publications were accepted, several simulators were developed and five startups were created.
The research topics were diverse:
- Solid mechanics with the simulation of the brain, the ear, the bones, the heart, the liver
- Fluid dynamics with the simulation of fat filling and blood flow in aneurysms
- Thermodynamics with thermo-ablation of tumors
- Many other topics as image processing, animation or biological applications
Based on an advanced software architecture, SOFA allows to:
- Create complex and evolving simulations by combining new algorithms with existing algorithms
- Modify most parameters of the simulation (deformable behavior, surface representation, solver, constraints, collision algorithm, ...) by simply editing a XML file
- Build complex models from simpler ones using a scene graph description
- Efficiently simulate the dynamics of interacting objects using abstract equation solvers
- Reuse and easily compare a variety of available methods
- Transparently parallelize complex computations using semantics based on data dependencies
- Use new generations of GPUs through the CUDA API to greatly improve computation times
A key aspect of SOFA is the use of a scene graph to organize and process the elements of a simulation while clearly separating the computation tasks from their possibly parallel scheduling. The description of a SOFA simulation can easily be done in an XML file. For even more flexibility, a Python plugin allows to script simulations using the Python language.
Basically, a SOFA scene-graph is composed with:
- Nodes: used to categorise the components and keep the XML file clean (mechanical node, collision node, visual node, ...)
- Components: main elements used to build a scene (solver component, forcefield component, rendering component, ...)
- Datas: everything that components have to deal with (forces, velocities, positions, ratios, ...)
Solvers & Models
The SOFA engine provides many solvers and physical models:
- Physical models:
All types of topology are available in SOFA (points, edges, triangles, quads, tetrahedra, hexahedra). Loaders implemented in the public version of SOFA allow to load various type of formats (obj, msh, vtk, etc.). Using these loaders, your mesh is directly uploaded in the simulation. Finally, SOFA proposes algorithms handling topological changes.
Different methods are implemented in SOFA to build a simulation involving collisions:
- Collision models: spheres, triangular meshes, distance fields (preliminary); with AABB-tree or octree bounding volume hierarchies
- Collision detection methods: proximity, continuous (preliminary)
- Collision response methods: (implicit) penalties, LCP-based constraints
The SOFA architecture relies on a multi-model representation which allows to have several representations (e.g. mechanical, thermal and visual) of the same object. Those different representations are connected together through a mechanism called Mapping. With this features, it is also possible to have models of very different nature interacting together (e.g. rigid bodies, deformable objects and fluids).
To extend its capacities and provide more features, SOFA is bundled with a lot of plugins:
- Drivers for VR / haptic / simulation devices (Geomagic®, ARTTrack™, Novint® Falcon™…)
- Visualization and simulation of medical images
- Python scripting
- GPU computing using the CUDA API
Organized each year, the SOFA Day is a one day event dedicated to SOFA. This event is open to everyone interested in SOFA, from beginner to advanced users. It contains an introduction to SOFA, several tutorials (adapted to the audience) and a large time to experience SOFA with the help of the instructors.
- Represent the identity of SOFA
- Organize and develop the community
- Distribute and make SOFA more stable
- Allard, Jérémie; Cotin, Stéphane; Faure, François; Bensoussan, Pierre-Jean; Poyer, François; Duriez, Christian; Delingette, Hervé; Grisoni, Laurent (2007). "SOFA - an Open Source Framework for Medical Simulation". Medicine Meets Virtual Reality (MMVR'15). Retrieved 24 January 2017.
- "SOFA Features".
- "SOFA Consortium".