Last month, the last commercial Boeing 747 was delivered to cargo operator Atlas Air. Present at an event marking this final delivery was Boeing CEO Dave Calhoun, who told Bloomberg TV that “Autonomy is going to come to all of the airplanes eventually […] The future of autonomy is real for civil”.
Known in the aviation industry as “Queen of the Skies”, the 747 revolutionized air travel as the world’s first twin-aisle passenger jet, and became paradigmatic of the age in which air travel became commonplace. It seems fitting that, with the twilight of one trailblazer, the arrival of the next revolutionary aircraft would be heralded.
In reality, the arrival of autonomous aircraft is far beyond a speck on the horizon. Autonomous planes have in fact spent decades in U.S. airspace, largely in the form of military property.
It’s clear that the aviation industry believes strongly that autonomous flight is the future, with companies from behemoth to startup size investing in the blossoming market. Boeing rival Airbus’ has an entire subsidiary dedicated to testing autonomous pilot assistance technology known as DragonFly, which head demonstrator Isabelle Lacaze explains “enable[s] an aircraft to ‘see’ and safely maneuver autonomously”. Companies such as Xwing have been valued at hundreds of millions of dollars without ever having designed a non-autonomous aircraft; around $7 billion has been raised for startups alone.
A decade ago, the prospect of self-flying commercial planes was a theoretical, rather than practical, discussion. Now, a large portion of the aviation industry believes that small autonomous planes will be carrying passengers by the end of this decade.
The main obstacle to a pilotless jumbo jet full of passengers is, well, the pilotless jumbo jet full of passengers. Dave Calhoun recognized in the same interview that public perception warming to the idea of a plane without a human qualified to save its passengers from a disastrous nosedive would “take time. Everyone’s got to build confidence. We need a certification process that we all have faith and believe in.”
A 2018 Ipsos public survey found that only 15% of the U.S. public would feel safe on a completely pilotless plane, whilst 81% of those surveyed said they would be strongly opposed to traveling on an autonomous aircraft. When offered hypothetical reductions to ticket prices, few changed their stance; the most aggressive discount of 30% still encountered a 66% rate of refusal.
It’s likely that time and the aviation industry’s professed trust in autonomous planes will sway the public. A sticking point is likely to be pilots and their unions, who realize that pilotless flight poses an existential threat to their livelihoods.
Unsurprisingly, many in the pilot unions are of the opinion that the impending revolution is motivated by money and gimmicks – views that will not encourage the public to trust pilotless flight and, significantly, will extend the shelf life of civil aviation pilots. A 2016 UBS report estimated that autonomous aircraft could save the cargo and passenger industry around $35 billion a year, but also acknowledged that 63% of respondents to an accompanying survey said they would not fly in a pilotless plane.
The discrete facts that airlines will save money by removing pilots and that the public majority does not trust planes without pilots have been combined by some to suggest that autonomous flight increases airline profits and reduces passenger safety. Dennis Tajer, spokesman for the Allied Pilots Association, which represents 15,000 American Airlines pilots, insists “It’s all about money […] Manufacturers are looking for the next innovative technology […] and airlines are looking at how they can do this more cheaply.”
Overlooked by many is the fact that it is more likely that the FAA would relax the current legislation of two pilots per civil flight to one, rather than none, leading to cheaper flights (for airlines and for passengers) that would still retain the safety of a human pilot. Also underreported is that the first generation of autonomous aircraft, at the very least, will be flown by a pilot; the difference will be that these pilots will be on the ground, flying aircraft remotely.
In an interview with Axios, aerospace specialist Jay Carmel explained that “when we say fully autonomous, it’s not as if these planes will have no humans involved”. Avascent, the consultancy firm for whom Mr. Carmel works and which conducted a thorough study into the future of self-flying aircraft, concluded that “it’s really a shift in how humans are interacting with the plane. The human is going to be much more of an observer, in a management state”.
A major misconception is that self-flying planes will not be self-flying. Instead, the software for aircraft to make better decisions is the primary focus of current research. As of 2023, roughly 90% of flying on commercial passenger flights is already on autopilot. As co-author of the Avascent study, Josh Pavluk, observed, “There’s nothing ‘unmanned’ about autonomous flight”.
The same AIA and Avascent study estimated that autonomous aviation would generate 100,000 jobs, many of them in engineering and technological support, by 2040. The market for self-flying planes, reports Avascent, will grow 25% annually to a market of $325 billion by the same date.
While it may take some time for the public to come to terms with being passengers on pilotless planes, it is likely that the majority would willingly accept more cheaply delivered goods from autonomous cargo planes. FAA regulations dictate that, until autonomous technology has been certified by U.S. law, one safety pilot will have to be on board. The aforementioned startup Xwing has prioritized cargo flight, taking the most widely used airframe (the Cessna 208 Grand Caravan) in the U.S. cargo industry; CEO Marc Piette was direct about their reasons for choosing cargo, as “the best first place to deploy this. And we’ve been very deliberate.”
For now, Xwing is running test flights, whereby a flight’s coordinates are pre-programmed, a human pilot sits in the cockpit, and a human controller operates the Cessna from the ground and interfaces with air traffic control. “It’s really a one-click thing, you engage the system and it runs its mission,” says Piette; “The safety pilot can disconnect a system and revert the aircraft to manual flying, but otherwise doesn’t do anything but monitor the system. It’s a very boring job”. Critical to Xwing’s mission is that they are not pushing for, or depending on, changes to aviation regulations – chief compliance and quality officer Earl Lawrence explained in an interview with Forbes that Xwing’s goal was “bringing this category of aircraft to the cargo market [without] changing the rules. We are following the regulations”.
Xwing’s approach may be the model to follow. By flying within the regulations while testing autonomous systems, cargo airlines can accumulate a body of evidence in the form of their flight logs that demonstrates that self-flying aircraft are cost-effective and safe, without creating contention with the FAA by pushing for legislative change. Given no risk to passenger life, it is probable that requirements for a pilot in the cockpit of cargo planes will be relaxed before any change to commercial passenger regulation. Once self-flying cargo planes are in the sky in numbers, it is likely perception within the aviation industry will become more confident, and the public will begin to reconsider their stance towards autonomous commercial planes.