Wes Bethel, Lawrence Berkelay National Laboratory, Information
and Computing Sciences Division
Janet Jacobsen, Lawrence Berkeley National Laboratory, Earth
Andy Austin and Mark Lederer, BP Exploration
Todd Little, Western Atlas Software
VR technology may be thought of as a user interface. It is useful to think of a spectrum, ranging the gamut from
command-line interfaces to completely immersive environments. In the former, one uses the keyboard to enter
three- or six-dimensional parameters. In the latter, three- or six-dimensional information is provided by trackers
contained either in hand-held devices or attached to the user in some fashion, e.g. attached to a head-mounted
display. Among the advantages of the former, rich, extensible and often complex languages are a vehicle
whereby the user controls parameters to manipulate object position and location in a virtual world. In the latter,
the user interacts with these parameters by means of motor skills which are highly developed.
Two specific geoscience application areas will be highlighted. In the first, we have used VR technology to manipulate three-dimensional input parameters, such as the spatial location of injection or production wells in a reservoir simulator. In the second, we demonstrate how VR technology has been used to manipulate visualization
tools, such as the position of a "rake," presented to the user in the form of a virtual injection well, which is used
in computing streamlines through fluxes.
We will discuss some issues to be considered when implementing VR and will describe our software development and user-working environments. Much of our work has been to create a software infrastructure to support
VR. This infrastructure is a collection of software "building blocks" and a visual programming interface for constructing a complete "program" for visualization. Our scientists can rapidly and easily interface several different
VR input devices, such as a Spaceball or magnetic trackers, to a variety of computational or visualization tools.
Our work on interface and visualization issues led us into two technical areas which will receive elaboration in
this paper. One of these is an optimized space-search algorithm which can be used by many scientific visualization tools, such as a streamlines solver. Another is our experience in making practical use of magnetic trackers as
an input device.
Will will conclude with a summary of the motivation for this work, and give an example of near-future extensions to this work.
Lawrence Berkeley National Laboratory
Mail Stop 50-F
Berkeley, CA 94720
phone: (510) 486-7353
fax: (510) 486-5548