AI pilot can navigate crowded airspace

According to the team, the artificial intelligence can safely avoid collisions, predict other aircraft’s intent, track aircraft and coordinate with their actions, and communicate over the radio with pilots and air traffic controllers. The researchers aim to develop the AI so the system’s behaviours will be indistinguishable from those of a human pilot.

“We believe we could eventually pass the Turing Test,” said Jean Oh, an associate research professor at CMU’s Robotics Institute (RI) and a member of the AI pilot team, referring to the test of an AI's ability to exhibit intelligent behaviour equivalent to a human.

To interact with other aircraft as a human pilot would, the AI uses both vision and natural language to communicate its intent, whether piloted or not. This behaviour leads to safe and socially compliant navigation. Researchers said they achieved this implicit coordination by training the AI on data collected at the Allegheny County Airport and the Pittsburgh-Butler Regional Airport that included air traffic patterns, images of aircraft and radio transmissions.