Semi-automated Simulation Transformation for DDDAS

David Brogan, Paul Reynolds, Robert Bartholet, Joseph Carnahan, and Yannick Loitière

Computer Science Department, University of Virginia, Charlottesville, VA 22901, USA
dbrogan@virginia.edu
pfr@virginia.edu
bartholet@virginia.edu
carnahan@virginia.edu
ycl2r@virginia.edu

Abstract. Ultimate DDDAS success demands that DDDAS simulations be increasingly reconfigurable and adaptable to a growing variety of runtime sensor feedback. Because we expect a simulation’s requirements to change during its lifetime, a new emphasis is placed on designing simulations that are prepared for transformation. In this paper, we address this new interest in designing for transformation. Our technology combines the specialized insight of simulation designers with the principled application of automation techniques to capitalize on untapped human and computational resources. In addressing simulation transformation issues that arise at design time, composition time, and runtime, we demonstrate how a semi-automated process impacts the entire simulation life cycle. The resulting suite of simulation transformation tools supports the crosscutting needs of DDDAS practitioners.

LNCS 3515, pp. 721-728.

Last modified: