XQ-58A 'Valkyrie' drone successfully flown by AI pilot

The Air Force Research Laboratory has announced that an XQ-58A "Valkyrie" drone has successfully been flown entirely under AI control for around three hours.
Christopher McFadden
XQ-58A "Valkyrie" demonstrator, circa 2019.

88 Air Base Wing Public Affairs/Wikimedia Commons 

An experimental XQ-58A "Valkyrie" drone has officially been flown under artificial intelligence control, the Air Force Research Laboratory (ARFL) announced. Conducted on July 25 at the Eglin Test and Training Complex in Florida, the test flight saw the drone entirely AI-controlled for around three hours.

This test follows around two years of research and development between a partnership with Skyborg Vanguard, a team made up of personnel from the Air Force Research Laboratory, and the Air Force Life Cycle Management Center with the intent of creating unmanned fighter aircraft.

AI loyal wingman

In case you are unaware, the Kratos XQ-58 "Valkyrie" is an unmanned combat aerial vehicle (UCAV) that Kratos Defense & Security Solutions are currently testing for the United States Air Force's Low-Cost Attritable Strike Demonstrator (LCASD) program. The "Valkyrie" was designed to be stealthy and was built as part of the USAF Research Laboratory’s Low-Cost Attritable Aircraft Technology (LCAAT) project portfolio.

“This sortie officially enables the ability to develop [artificial intelligence and machine learning] 'agents' that will execute modern air-to-air and air-to-surface skills that are immediately transferrable to the CCA program,” said Col. Tucker Hamilton, the chief of AI test and operations with the Air Force. The CCA program, also known as collaborative combat aircraft, aims to develop combat drones that can function alongside human-piloted F-22 and F-35 aircraft (aka "Loyal Wingmen").

The Autonomous Air Combat Operations (AACO) team at the lab developed algorithms for the flight. They spent millions of hours perfecting these algorithms through simulations, sorties with the X-62 VISTA experimental aircraft, working with the XQ-58A, and conducting ground test operations.

“The mission proved out a multi-layer safety framework on an AI/ML-flown uncrewed aircraft and demonstrated an AI/ML agent solving a tactically relevant “challenge problem” during airborne operations,” said Col. Hamilton.

Past XQ-58A "Valkyrie" flights have supported the Air Force's research into loyal wingmen. These completely autonomous aircraft will fly with and support human pilots and 6th-generation combat aircraft during sorties.

“AI will be a critical element to future warfighting and the speed at which we’re going to have to understand the operational picture and make decisions,” Brig. Gen. Scott Cain, the lab’s commander, said in the announcement. “AI, Autonomous Operations, and Human-Machine Teaming continue to evolve at an unprecedented pace, and we need the coordinated efforts of our government, academia, and industry partners to keep pace," he added.

Vital for DoD

“AACO has taken a multi-pronged approach to uncrewed flight testing of machine learning Artificial Intelligence and has met operational experimentation objectives by using a combination of High-performance computing, modeling and simulation, and hardware in the loop testing to train an AI agent to safely fly the XQ-58 uncrewed aircraft,” said Dr. Terry Wilson, AACO Program Manager.

The Department of Defense (DoD), AFRL explained, is dedicated to the responsible use of AI. To achieve this, developers and users of AI-enabled autonomy must collaborate with acquisition specialists.

Add Interesting Engineering to your Google News feed.
Add Interesting Engineering to your Google News feed.
message circleSHOW COMMENT (1)chevron
Job Board