CUDA parallel implementation of airspace conflict detection
Abstract
The maintaining of separation between aircraft or any air bound projectile is an important concept in the daily lives of many people. Whether it is air traffic controllers in public transportation or military use in the battlefield, maintaining projectile separation can be the difference between success and tragedy. To prevent and resolve failures and operational errors that may occur, automated tools are being developed to assist in controlling traffic in airspaces. These automated tools not only need to process logistical data for each aircraft or projectile in a particular airspace, but they also need to provide collision-avoidance techniques such as airspace deconfliction. Developing these kinds of automated tools may be too computationally demanding for typical Central Processing Units (CPUs). Solutions to handling such computational stresses may come in the form of supercomputers or a series of computers executing in parallel. Though these come at a price and require additional resources, a solution involving low cost Graphical Processing Units (GPUs) may provide the processing power required by the air traffic control systems because of the parallel nature in which it is able to perform computations.
Degree
M.S.E.
Advisors
Thompson, Purdue University.
Subject Area
Computer Engineering|Computer science
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.