Artificial intelligence’s (AI) impact on the business aviation industry promises to be profound, encompassing virtually every aspect of its operations. The different applications and their implications were threaded throughout the convention floor and industry panels during the 2024 edition of NBAA-BACE.
“Artificial intelligence has transformed from the realm of science fiction to becoming an integral part of our daily lives and industries,” said Todd Giles, Honeywell Aerospace's chief technology officer at the OEM’s annual forecast event on Sunday evening, unofficially kicking off NBAA-BACE.
From enhancing pilot decision-making to increasing productivity across business functions, AI’s influence is growing rapidly, with perhaps the most profound changes coming in the safety sector.
In an NBAA-BACE panel discussion around AI first-movers on Tuesday morning, Reliable Robotics co-founder and CEO Robert Rose cited a joint study from NASA, the FAA, and the University of Tulsa that examined common causes of accidents in small aircraft and determined that 68% of those accidents could be prevented with advanced automation systems. He added that the top causes of fatal accidents in small aircraft are controlled flight into terrain (CFIT) and loss of control in flight.
“Our first certification is a continuous autopilot engagement system, an always-on autopilot is targeted in preventing those first two causes of fatal accidents in small aircraft.” In addressing CFIT, he explained, “You’ve got an autopilot that is tightly coupled with your flight management system, and it’s crosschecking your flight path to make sure you are keeping a safe distance from terrain and other aircraft.”
Rose noted the AI autopilot would not allow the aircraft to exit its operating envelope. “Then loss of control is no longer a thing, so you’ve got roughly a third of fatal accidents in small aircraft are eliminated right there.”
“Eventually, we're going to put that technology into aircraft and enable you to remotely pilot it so that the pilot is on the loop rather than in it,” said Rose. “2027 is when we are targeting having our first operations of our remotely piloted system for the Cessna Caravan,” he said.
“I think the big misconception is we're not looking to eliminate humans or eliminate pilots. It's just about moving the pilot out of the cockpit and putting them in a control center and allowing them to do their job more effectively,” Rose added.
AI’s role in assisting pilots during flight is also coming into focus. Trung Pham, chief scientist for AI and machine learning at the FAA, noted that while AI hasn't yet reached the point of fully operating aircraft, it is already making strides in augmented intelligence. “With an AI system, the system can see more than what we can focus on and inform us in a role of monitoring what's going on—and inform us of certain precursors that can lead to accidents or incidents,” he said.
“We think the human is going to be part of this equation for a long time—there are things humans do just naturally that are really hard to do,” said Jon Damush, president and CEO of uAvionix. He said the question is how to use AI to simplify the cockpit, separate out menial chores, and allow them to concentrate on the hard tasks.
Merlin, backed by Google, is developing a nonhuman pilot system to serve as an autonomous third "pilot" in the cockpit, with the goal of eventually enabling single-pilot operations in most aircraft. The company is already working with the U.S. military to reduce crew size on aircraft like the C-130J and KC-135R.
AI technologies such as computer vision are also enhancing the ability to inspect aircraft components that are otherwise difficult for humans to assess, thus improving safety and efficiency. Damush said his company is working on a ground-based aircraft detection system to help pilots recognize threats. “There is no data set out there for that; we had to go get it,” he told the audience.
“We literally flew airplanes at each other and at our ground stations over 17,000 times, and then we built some specialized tools for humans to go in and say that’s a Cessna, that’s a helicopter, that’s a bird, etc.” That data is uploaded to the system which then finds all the permutations of the images and builds classifiers.
Likewise, Daedalean is building perception systems that are designed and integrated with artificial intelligence. “What we bring is what we call situational intelligence explained Yemaya Bordain, the company’s president for the Americas.
“What a human does is anticipate and react to future threats,” Bordain said. “We are taking our traffic detection algorithm and we're implementing it in a traffic collision avoidance system,” essentially acting as a second set of eyes for the pilot. Daedalean’s system identifies “non-cooperative” traffic, such as birds, drones, hot air balloons, or even paragliders on the traffic display along with any aircraft with a transponder. She said Daedalean expects to introduce this technology into the market within the next year and a half.
Bordain noted that Daedalean owns a small fleet of aircraft that it constantly flies to collect data and refine its database to improve its AI system accuracy. “We keep all of that in-house, and it's probably the most valuable asset within our company.”
AI is extending to the maintenance realm as well. Pratt & Whitney recently launched an AI-based aircraft engine analysis tool for geared turbofan and V2500 powerplants. Called Percept, the inspection tool combines computer vision technology with a video intelligence operating system developed by Indian AI start-up Awiros.
The company's maintenance technicians use a cloud-based interface on a mobile device to capture images of an engine, and the app provides real-time parts availability status. According to Pratt & Whitney, Percept has reduced the time it takes to conduct an engine inspection by nearly 90%.
In another example, Honeywell Aerospace recently began using AI and machine learning to help manufacturers and MROs modernize their processes and save costs by leveraging data to inform critical decisions. Honeywell unveiled the cloud-based software solution, called Forge Performance+, in April as an addition to its Forge suite of digital tools for aerospace manufacturers and MROs.
“If you go to any one of our websites, you'll see that we are demonstrating capabilities that have never been done before,” Bordain concluded. “They are beyond what anyone thought could exist. So it's happening right now, and each one of us is working on how we can certify those systems and ensure that they're safe.”