9/9/2020

AI Bests Human Fighter Pilot in Alpha Dogfight Trial at Johns Hopkins APL

Read full story
Article Keywords
Trustworthy Systems
Human-Machine Teams
Machine Learning
Reinforcement Learning



Collaborating ISC Lab(s)

UxV Gym

In a matchup of human versus machine, the decisive winner was Heron Systems’ artificial intelligence (AI) against an experienced human F-16 fighter pilot in a simulated aerial battle that capped the AlphaDogfight Trials (ADT) on Aug. 20 at the Johns Hopkins Applied Physics Laboratory (APL) in Laurel, Maryland. The competition took place at the APL’s Intelligent Systems Center. 

 

Viewers watched on YouTube as Heron Systems, a Maryland-based defense contractor that builds autonomous agents and AI-powered multi-agent systems, went 5–0 against F-16 pilot “Banger” in the Defense Advanced Research Projects Agency’s (DARPA) virtual competition. ADT, part of DARPA’s Air Combat Evolution (ACE) program, was intended to forge greater trust in autonomous systems for air combat. 

 

APL’s AI agents gave teams a run on the first day with all eight teams virtually flying their algorithms against five APL-developed computer adversaries. APL’s algorithms were developed to perform at different skill levels, allowing the industry agents to train similarly to actual pilots.

 

Article Keywords
Trustworthy Systems
Human-Machine Teams
Machine Learning
Reinforcement Learning