Accelerator for UAV Modeling in Near-Ship Environments Based on Commodity Graphics Cards

UAVS (Unmanned Aircraft Vehicle System) integration with naval vessels is currently realized in limited form. This is largely due to the fact that the operational envelopes of these vehicles are based solely on at-sea flight testing. In addition to the complexities involved with at-sea flight testing, the unsteady nature of ship-airwakes and the use of automated UAV control software necessitates that these tests be extremely conservative in nature. Instead of flight testing, modeling and simulation could be used to predict UAV operation under these conditions. Unfortunately, the computational requirements for a fully-coupled computational fluid dynamics (CFD) solution render such an approach impractical. To overcome this limitation requires the creation of simulators that model the full level of detail required but have drastically reduced run times. To overcome these obstacles requires a two-pronged approach: algorithmic improvements and implementation improvements. In this project, we address both of these needs. We will create a solver by coupling the combined computational aeroacoustic (CAA) method with more traditional CFD algorithms and implement it on commodity graphics cards to reduce computation times by orders of magnitude. The result will be a tool for the rapid and accurate simulations analysis of UAVs in near-ship environments.