AFRL Demonstrates AI-piloted XQ-58A Valkyrie Uncrewed Jet Aircraft

 https://youtu.be/OrYvAUgF6vM

AFRL Demonstrates AI-piloted XQ-58A Valkyrie Uncrewed Jet Aircraft U.S. Air Force Research Laboratory (AFRL) conducted first-ever flight of XQ-58A Valkyrie uncrewed jet aircraft using artificial intelligence agents (algorithms) on July 25. The mission demonstrated a multi-layer safety framework for AI/ML-flown uncrewed aircraft and showcased an AI/ML agent solving a tactically relevant problem during airborne operations. AI/ML agent's capabilities include executing modern air-to-air and air-to-surface skills, which can be transferred to the CCA (Comprehensive Combat Airspace) program. Algorithms controlling the XQ-58A were developed by AFRL's Autonomous Air Combat Operations team. AI algorithms underwent significant development and training, including millions of hours in high fidelity simulation events, sorties on the X-62 VISTA, Hardware-in-the-Loop events with the XQ-58A, and ground test operations. The team utilized a combination of high-performance computing, modeling and simulation, and hardware-in-the-loop testing to train the AI agent for safe uncrewed flight operations. The Kratos XQ-58 Valkyrie has a capacity of 600 kg internally and externally, with a length of 30 ft and a wingspan of 27 ft. It weighs 2,500 kg empty and can take off at a maximum weight of 6,000 kg. Valkyrie can fly at speeds of up to Mach 0.72, its range is 3,000 nautical miles while service ceiling is 45,000 ft. The $4-5 million drone can join drone swarms and launch from various platforms