The United States Air Force demonstrated AI-enabled (Artificial Intelligence) air combat during a successful launch of an Unmanned Aerial Vehicle XQ-58A Valkyrie.

According to an announcement from the Air Force Research Laboratory (ARFL), the said flight shall assist in developing a trained tactical autonomy algorithm from simulation through flight test on a high-performance Unmanned Aerial Vehicle (UAV). AI algorithms, developed and trained by the Air Force Research Laboratory’s Autonomous Air Combat Operations, were integrated into an XQ-58A and flown in the Eglin Gulf Test and Training Range.

Trained through deep reinforcement learning, the AI algorithms used neural networks to fly the UAV against simulated opponents using simulated mission systems and simulated weapons.

Also read: AQUILA | MBDA to lead consortium for European interceptor against hypersonic threats

AI testing requires combining new and conventional test and evaluation techniques, while the team has a lot of lessons learned that will be used to inform future programs, said Ryan Bowers, lead test engineer for the effort.

The flight test, executed by the 40th Flight Test Squadron and supported by AFRL and Kratos Unmanned Aerial Systems, was a continuation of the successful July 25 test flight. The previous flight presented an AI-enabled, high-performance, uncrewed air vehicle for the Department of Defence (DoD) and demonstrated standard aviation tasks, navigation tasks, and safety guardrails for risk mitigation and safety build-up.

The Pentagon is committed to the responsible employment of AI technologies. To achieve responsible use of AI requires teaming of developers and users of AI-enabled autonomy working in collaboration with acquisition specialists.

AI will be a critical element to future warfighting and the speed at which operators will have to understand the operational picture and make decisions. 

Also read: Hermeus | Full-Funded Quarterhorse with $60 Million US Air Force Partnership – VIDEO