This first appeared as a Comment in the 17 December issue of Flight International.
A senior regulator at EASA crystal-ball gazing about the prospect of pilotless – or at least remotely-piloted – airliners says: “It may happen in 50 years, but I won’t bet on it.”
All experts believe the automation of the pilot’s role is inevitable. The argument is about how soon and what the technical milestones will be.
Meanwhile, automation proceeds apace in the real world, and it is an undisputed fact that it is de-skilling pilots. In some notorious recent cases it has rendered them unable to cope when the automation failed.
However, a crucial piece of evidence about whether pilots are essential is missing. The world’s aviation authorities have all the data they need about how often pilots fail and passengers die as a result, but they do not collect data on pilot successes.
More important than the spectacular incidents like the Hudson River ditching and Qantas QF32 are the countless times that things do not go exactly as planned, yet the pilots cope effortlessly and the occurrence is not even worth logging.
Most of the adaptive tasks that pilots perform on a routine flight are unremarkable. However, if the industry is even toying with the idea of pilotless airliners, the minutiae of a flightcrew’s work must first be better understood. Only then can a computer’s suitability to replace flesh-and-blood pilots be properly assessed.