Computational mechanics routines, in other terms physical engines, are employed massively in engineering and entertainment industries, being computer games, movies, or computer-aided design applications. However, current open-source and even commercial libraries either lack speed or accuracy. We aim at both. Finite Elements are an accurate methods for solving differential equations, such as elasticity or fluid-solid interactions.
Currently, we are writing a set of Finite Element softwares able to grasp some mechanical properties of various materials, in particular elastoplastic ones such as CNT (Carbon Nanotubes) materials with a non-linear plasticity model, and fluids with Navier-Stokes equations coupled with temperature and particle mechanics.
Usually simulations require a lot of CPU-hours, and our objective is to reduce drastically the required time by employing smart data structures and algorithms, and employing GPUs for accelerating the solution process (e.g., OpenCL, CUDA).