Unmanned Aerial Vehicles (UAVs) have assumed several roles once dedicated to manned platforms. Predator and Global Hawk UAVs distinguished themselves in the recent Iraqi conflict by delivering radar- and electrooptical/infrared (EO/IR) intelligence from the battlespace without exposing human pilots to risk. These UAVs employ commercial off-the-shelf (COTS) processing systems that have been hardened for shock, vibration, rapid temperature changes and corrosive environments, and are readily available on the market.
The strict constraints on the size, weight and power available for a UAV sensor processing subsystem require high-density computing solutions. In today's most widely deployed UAV—the U.S. Air Force's Predator built by General Atomics — a COTS-based signal processor comprising several multiprocessors interconnected by an ANSI/VITA standard RACEway switch fabric, processes the unrelenting, highvolume stream of synthetic aperture radar (SAR) data. Similarly, the higherflying, longer-duration Global Hawk UAV, built by Northrop Grumman (www.northgrum.com), also carries a high-performance SAR system with a signal processing subsystem based on the RACEway switch fabric multicomputer architecture.
Global Hawk will require increasing amounts of processing throughout its service life, as mission planners look for ways to pack more capabilities into its large yet finite airframe. As an example, designers are striving to move more signal and image processing from ground stations to the UAV itself, in order to process more data in real time than can be down-linked from the aircraft to a ground-based system in a constant stream. The ability to process the data in-flight also opens new mission capabilities, such as passing the results to tactical aircraft, in near real time, for use in targeting or other functions.
View Entire Paper | Previous Page | White Papers Search
If you found this page useful, bookmark and share it on: