Inspenet, August 19, 2023.
The United States Air Force announced the successful flight of a drone equipped with artificial intelligence (AI) and machine learning systems, thus demonstrating the effectiveness of incorporating combat technologies into its air and space operations.
According to the United States Air Force Research Laboratory, the aircraft XQ-58A Valkyrie successfully completed a three-hour flight on July 25, 2023, marking a milestone as the first flight using machine learning-trained AI algorithms on this type of air vehicle. In addition, this model demonstrated its ability to process the necessary information and improve the use of autonomous aircraft.
The technology behind the XQ-58A Valkyrie drone
Regarding the embedded technology, the artificial intelligence algorithms were designed by the Laboratory and refined through extensive simulations, test flights with the X-62 VISTA experimental aircraft and ground tests.
On the other hand, the purpose of this initiative lies in the creation of autonomous aircraft driven by artificial intelligence and machine learning systems , so that they can fly and assist human pilots in making quick decisions during combat situations.
About the test
The evaluation units carried out the flight at the test and training facility in Eglin (United States). This trial is the product of a four-year collaboration between the Skyborg Vanguard and Autonomous Aircraft Experimentation (AAx) programs.
Colonel Tucker Hamilton, head of Artificial Intelligence Testing and Operations, told the Department of the Air Force that “the mission tested a multi-layered security framework and demonstrated that an agent solved a tactically relevant ‘challenge problem’ during airborne operations. This output officially enables the ability to develop agents who will execute modern air-to-air and air-to-surface skills that are immediately transferable to other autonomy programs.”
Meanwhile, Terry Wilson, AACO Program Manager, adds that “AACO has taken a multi-pronged approach to machine learning artificial intelligence unmanned flight testing and has met operational experimentation objectives by using a combination of High performance, modeling and simulation, and in-circuit hardware testing to train an AI agent to safely fly the XQ-58 drone.”