In the two weeks since the crash of Germanwings Flight 9525, there has been no lack of commentary about the failures and future of piloted aviation — much of it frustratingly ill informed. Planes are already so automated, consensus has it, that it’s only a matter of time before pilots are engineered out of the picture completely. And isn’t the best way to prevent another Andreas Lubitz to replace him with a computer, or a remote pilot, somewhere on the ground? Cue the aeronautics professor or university scientist who will blithely assert that yes, we are well on our way to a pilotless future.
The problem with this line of thought is that it begins with a false premise: the idea that jetliners today are super-automated machines whose pilots serve merely as backup in case of an emergency. Indeed, the notion of the automatic airplane that “flies itself” is perhaps the most stubborn myth in all of aviation. According to one recent story, pilots on a typical flight spend as little as seven minutes manually piloting their planes. Another story had it at three minutes! Statements like this are highly misleading.
True, these days pilots spend only a short amount of time with their hands on the control column or stick. But that does not mean we aren’t controlling the airplane throughout the entire flight. Our hands might not be steering the airplane directly, as would have been the case in decades past, but almost everything the airplane does is commanded, one way or the other, by the crew. The automation only does what we tell it to do. On the 767 that I fly, there are multiple ways to set up and command any routine climb, descent or change of course. Meanwhile, more than 99 percent of landings, and a full 100 percent of takeoffs, are performed manually.
People might be surprised at how busy a cockpit can become on the most routine flight, even with all of the automation running. There are stretches of low workload during which, to the nonpilot observer, it would seem that very little requires the crew’s attention. But there are also periods of very high workload.
The other day I piloted a flight from the Caribbean to New York. We had bad weather the whole way, followed by a low-visibility approach into Kennedy Airport. The autopilot was on pretty much the entire time, but there were numerous altitude, course and speed changes to coordinate; holding, arrival and approach patterns to set up and fly; and all of the requisite communicating with air-traffic control, company staff and cabin crew. By the end of the flight, my voice was hoarse.
The best analogy, I think, is modern medicine. Cockpit automation assists pilots in the same way that robots and other advanced medical equipment assist surgeons. It has improved their capabilities, but a plane no more flies itself than an operating room performs a hip replacement by itself.
Such wishful thinking is perhaps symptomatic of our infatuation with technology and gadgetry, and the belief that we can compute our way out of every problem. The proliferation of drone aircraft also makes it easy to imagine a world of remotely controlled passenger planes. In fact, Boeing has acquired a patent on a sophisticated, remotely operated autopilot system.
But for now these things exist only in the experimental stages. A handful of successful test flights does not prove the viability of a system that would carry up to four million passengers every day around the world. And remember that drones have wholly different missions from those of commercial aircraft, with a lot less at stake if one crashes.
More than 415 large drones flown by the American military have crashed in accidents since 2001, a record that is acceptable, if expensive, for remotely controlled aircraft, but that would be disastrous for civil aviation. A flight is subject to so many potential problems and contingencies. Even minor malfunctions can be complicated; the idea of trying to handle a serious emergency from a room thousands of miles away is about the scariest thing I can imagine.
And aside from the tremendous safety and technological challenges, we’d also need a more or less full redesign of our aviation infrastructure, from developing a fleet of highly expensive, currently nonexistent aircraft to a new air-traffic-control system. Each of these would cost tens of billions of dollars and take many years to develop. And in the end, you’d still need pilots to operate these aircraft, albeit from a remote location.
Many of you are thinking: Here’s this Luddite pilot who can’t bear the prospect of seeing his profession go the way of the Teletype operator. Sure, I’m biased. It’s also true that, unlike many of those who might counter my assertions, I have a solid understanding of the complexities of commercial flying, and of the complications that these futuristic endeavors would entail.
I’m less concerned about the future than I am about the present: Should pilotless air travel become a reality, so be it. Until then, however, we’re living in a world in which almost everybody travels by air, and the public deserves to have an accurate sense of how planes actually fly, and what we pilots actually do for a living.
Patrick Smith, is an airline pilot and the author of the book Cockpit Confidential: Everything You Need to Know About Air Travel.