The US Air Force’s XQ-58A Valkyrie experimental combat drone made its first artificial intelligence-assisted flight. On July 25, 2023, a robotic aircraft made a three-hour flight to the Eglin Testing and Training Complex in Florida.

The latest test is the result of the Skyborg Vanguard program, which aims to improve the AI/ML agent’s ability to fly safely while solving tactically relevant missions. The agent received simulator training, where he honed his skills during millions of hours of high-fidelity simulations and sorties in the X-62 VISTA with algorithms developed by AFRL’s Autonomous Air Combat Operations (AACO) group.

The joint project of Kratos Unmanned Aerial Systems (KUAS) and the US Air Force Research Laboratory (AFRL) XQ-58A Valkyrie aims to develop the technology necessary to create a jet-powered autonomous combat aircraft that can operate independently without human control. , together with other drones and as part of a human/machine team.

“The mission demonstrated a multi-layered security framework in an AI/ML-controlled unmanned aerial vehicle and demonstrated how an AI/ML agent solves a tactically important ‘challenge problem’ during air operations,” said Colonel Tucker Hamilton, DAF AI Testing and Operations. Chef. “This sortie formally enables the development of AI/ML agents that will use modern air-to-air and air-to-surface capabilities that can be instantly transferred to other autonomous applications.”

The idea is to reduce the ever-increasing cost of manned aircraft by providing the U.S. Air Force with a robotic alternative that can accompany and work with fighter jets and be deployed into particularly dangerous areas.

With a wingspan of 22 ft (6.7 m) and a top speed of 567 knots (652 mph, 1,050 km/h), the Valkyrie has an operating ceiling of 44,997 ft (13,715 m) and a range of 2,128 nautical miles (2,449 mi). , 3,941 km). It is capable of carrying up to eight weapons, including JDAMs and other small-caliber bombs.

“AI will be a critical element of future warfare and the speed with which we understand the operational landscape and make decisions,” said Brigadier General Scott Kane, AFRL commander. “Artificial intelligence, autonomous operations, and human-machine integration continue to advance at an unprecedented pace, and we need a coordinated effort from our government, academia, and industry partners to keep pace.” Source