You would be hard-pressed to find anyone who feels that flying couldn’t be made safer. The good news is that researchers are working to make flying safer than ever. Graduate students from Carnegie Mellon University’s Robotics Institute AirLab recently presented at a NASA symposium in Los Angeles. Their project, World2Rules, is a breakthrough AI designed to stop runway accidents before they can even start.
Jay Patrikar and Jack Wang are two students working in Sebastian Scherer’s AirLab at CMU’s Robotics Institute. They wanted to create a tool that gives pilots and air traffic controllers crucial extra minutes to avert disaster. To do this, they built a “neuro-symbolic” AI. While typical AI operates as an unreadable “black box,” this new method blends massive data-processing power with logic that humans can easily read, understand and double-check – a necessity for life-or-death aviation safety.
One challenge of using real data from airports and planes is the sheer volume – a massive 10-terabyte set of raw Federal Aviation Administration (FAA) data spanning two years across 42 U.S. airports. To ingest and use that data, the student-led team would need powerful resources. They turned to the U.S. National Science Foundation’s ACCESS program to get an allocation on the Pittsburgh Supercomputing Center’s Bridges-2 machine.
We basically collect all aircraft surface movement data from airports in the U.S. … And that stream is pretty intense. It’s, like, one megabit per second every day, 24-7, 365 … So it’s a ridiculously high amount of data … We used PSC to train large trajectory forecasting models, [which] was first of its kind … We don’t want to understand [that] a crash is happening. We want to predict if a crash will happen in the future.
–Jay Patrikar, CMU Robotics Institute
By leveraging Bridges-2, the team achieved up to a 43% improvement in learning accurate safety rules compared to traditional modeling approaches.
To explore how these graduate students created this AI, you can find in-depth details here: New AI Learns Airport Safety Rules, Helps Avert Devastating Collisions.
If you’re a student in need of compute resources for your project, you can get started with ACCESS here.
Resource Provider Institution(s): Pittsburgh Supercomputing Center (PSC)
Resources Used: Bridges-2
Affiliations: Carnegie Mellon University
Funding Agency: NSF
Grant or Allocation Number(s): CIS220044
The science story featured here was enabled by the U.S. National Science Foundation’s ACCESS program, which is supported by National Science Foundation grants #2138259, #2138286, #2138307, #2137603, and #2138296.