The aerospace industry is definitely making great strides towards the machine learning era. The U.S. Air Force Test Pilot School and the Defense Advanced Research Projects Agency (DARPA) formally recognize the recent breakthroughs that ushered in the era of machine learning in the aerospace industry by Robert J. Collier in 2023. Recognized as a finalist for the trophy.
Their collaboration led to recent breakthroughs in artificial intelligence algorithms using the X-62A VISTA aircraft as part of DARPA's Air Combat Evolution Program.
“The possibility of autonomous air-to-air combat has been imaginable for decades, but reality has remained a distant dream until now. In 2023, the X-62A will be one of the most important obstacles in combat aviation. This was a transformative moment, and it was all made possible by the breakthrough work of the X-62A ACE team.” Air Force Secretary Frank Kendall said. Secretary Kendall will soon be flying in the X-62A VISTA, where he will witness first-hand the AI in a simulated combat environment during a test flight scheduled at Edwards.
The team achieved a significant milestone in the aerospace industry. Less than a year later, they installed a live AI agent into his X-62A's system and demonstrated the first AI-to-human engagement within visual range, also known as air combat. To accomplish this feat, the team made over 100,000 flight-critical software changes over his 21 test flights.
The success of AI air combat has proven that non-deterministic artificial intelligence can be safely used in aerospace. The X-62A VISTA faced a manned F-16 aircraft over Edwards, first using defensive maneuvers to establish initial flight safety before switching to aggressive high-aspect nose-to-aircraft engagements. Aircraft approached to 2,000 feet at 1,200 miles per hour.
The first use of machine learning-based autonomy in flight-critical systems will serve as the foundation for future aerospace AI advances that will be safer and more reliable in commercial and defense applications.
“The X-62A is a great platform that can be used not only to advance the current state of research and testing, but also to prepare the next generation of test leaders. The industry can look to the results of what the X-62A ACE team has done as a paradigm shift when we demonstrate that this can be done safely and responsibly. , fundamentally changed the conversation.” said Col. James Valpiani, Test Pilot School commander.
Historically, machine learning has been prohibited in aerospace due to its high risk and lack of independent control. However, traditional autonomy has been in place for decades. The X-62A is flown by a safety pilot with an independent ability to disable the AI, but they were not required to activate the safety switch at any point during the dogfight over Edwards. This success is a testament to the reliability of the algorithms used in the X-62A.
“DARPA's ACE program manager, Lt. Col. Ryan Heffron, said: “For these algorithms to be used in real-world settings, they need to be reliable.”
Although air combat was the main test scenario, it was not the end goal. The X-62A ACE program was designed to test autonomous artificial intelligence systems in the air. Therefore, all the lessons learned in air combat scenarios apply to all tasks given to autonomous systems.
“It's very easy to look at the X-62A ACE program and think it's under autonomous control. It can also do dogfights in the air, but that's missing the point. We've started testing autonomous artificial intelligence systems in the air. Air combat was the problem to solve. All the lessons we're learning apply to any task that can be given to an autonomous system.” said Bill Gray, the school's chief test pilot.
This achievement marks an important milestone in the development of autonomous systems for aerospace applications.
The machine learning breakthroughs achieved by the X-62A VISTA project will continue to advance lessons learned into future record-breaking programs. The X-62A VISTA continues to serve a variety of customers for research purposes while providing important academic lessons for the next generation of test leaders.
The ACE program is the result of a strong collaboration between academia, government, and private industry. Government partners in the ACE program include the Air Force Test Center, Air Force Research Laboratory, DARPA, and Air Force Test Pilot School. Academic partners for the program include Johns Hopkins University and his MIT Computer Science and Artificial Intelligence Laboratory. Industry partners involved in ACE include Calspan Corporation, Cubic Corporation, EpiSci, Lockheed Martin Skunk Works, and Shield AI.
