| authors |
Hnat, B. and Chapman, S.C. |
| year |
2000 |
| title |
Visualisation of multiscale datasets in numerical models of complex systems |
| source |
CIDAC, Volume 2 Issue 4 November 2000, pp. 236-241 |
| summary |
Recently advanced visualisation techniques receive increasing attention among scientists and engineers. These techniques should not only allow a direct analysis of multidimensional structures to provide intiutive and quick comparison of experimental data with model results, but also yield quantitative measures to support or negate the model itself. A key phenomenology of complex systems is that they are multiscale, i.e. the structures of interest in the data are on the wide range of spatiotemporal scale. Here we examine a simple, yet generic, example, a sandpile (avalance model), the threshold coupled map lattice (CML). The emergent phenomenology is broadband, but is represented on a (finite) grid. To study the 3-dimensional evolution of the threshold CML model of a sandpile as a function of control parameters, a semi-immersive virtual reality (VR) environment has been used. Sandpile profiles are calculated in real time and are superimposed over the colour grid respresenting one of the control parameters. This represents a generic problem of visualising multiscale structures in datasets derived from non-linear systems. The software models were developed in OpenGL and Glut library and will ultimately be implemented in semi-immersive VR on ImmersaDesk. To achieve sufficient speed in real-time display and to ultimately facilitate computational steering, the full data set must be represented geographically in a compact form without loss of information. |
| keywords |
Semi-Immersive Virtual Reality, 3-Dimensional Visualisation, OpenGL, CML |
| series |
journal paper |
| references |
Content-type: text/plain
|
| last changed |
2003/05/15 21:23 |
|
