[an error occurred while processing this directive]
[an error occurred while processing this directive]
|
|
|
Raymond Spiteri Faculty of Computer Science Dalhousie University Halifax, Nova Scotia |
|
Ordinary differential equations provide one of the most powerful ways to model and simulate real-life systems such as those arising from robotics or physics-based modelling in computer graphics. Traditional numerical methods for solving ordinary differential equations are typically of two types: implicit or explicit. Explicit methods are easy to implement but are inefficient on stiff problems. Implicit methods are effective for stiff problems, but their implementation is not trivial. Recently there has been a lot of interest in hybrid solvers because many problems naturally have stiff and nonstiff parts. In this talk I will discuss how hybrid solvers can achieve the best of both worlds: I will show that hybrid solvers can outperform both of the traditional types of methods.
Dr. Raymond Spiteri obtained his B.Sc. (Hons.) in Applied Mathematics (Theoretical Physics) in 1990 from the University of Western Ontario. He obtained his Ph.D. in Mathematics (Applied Mathematics) in 1997 from the University of British Columbia. He then spent two years as a post-doctoral fellow at McGill University in the School of Computer Science with cross-appointments in the Department of Mathematics and Statistics and the Centre for Intelligent Machines (Mechanical Engineering).
Dr. Spiteri is presently an Assistant Professor at Dalhousie University, jointly appointed between the Faculy of Computer Science and the Department of Mathematics and Statistics. His research interests are in scientific computation, with particular interest in physics-based models for computer graphics and problems derived from robotics and engineering applications. A major part of his research concerns the development, optimization, and implementation of algorithms and software for scientific computation. [an error occurred while processing this directive]