USING GPU’S FOR AIR TRAFFIC MANAGEMENTJanuary 17, 2012
Legroom is tight, babies are screaming and the food supply is running low. Such are the horrors of being stuck on the tarmac. Thankfully, the folks at Silicon Valley-based Optimal Synthesis are working to advance air traffic flow management using CUDA GPUs.
By some estimates, the number of aircraft in our skies will double by 2025. This is not news to the U.S. government, which started the NextGen multi-agency initiative in 2003 to use advanced computer algorithms and automation to assist human air traffic controllers.
In this two-part series, we’ll take a look at two companies using GPU computing to help solve tomorrow’s air traffic woes.
Our first story is about Silicon Valley-based Optimal Synthesis, which is working with NASA to dramatically accelerate air traffic flow models using GPUs and CUDA. Dr. Monish Tandale says that “today’s air traffic management is mainly human-centric,” but future air traffic volume will “use automation to allow more people to fly without being inconvenienced by congestion and delayed flights.”
The key to Optimal Synthesis’s work is the ability to analyze air traffic in real-time. “GPU computing allows us to exploit the parallelism in the trajectory prediction process,” explains Dr. Monish Tandale. “This in turn allows us to achieve real-time performance and analyze models with greater complexity.”
Using CUDA and GPU computing, Optimal Synthesis and NASA were able to achieve a 250X speedup over NASA’s baseline software.
Watch the video below to learn more.
Read the full interview with Dr. Tandale here.
We’ll follow up later with a conversation with Bart Gallet of Mosaic ATM, of Leesburg, Virginia, about his firm’s work in the field of “dynamic airspace configuration,” which decomposes the airspace into sectors that can be managed by a single controller.