Rockwell Collins has been tapped to lead a NASA-sponsored study into "single pilot" operations of commercial airliners.There has been a running joke for decades in the aviation industry, back to the days when the Flight Engineer's job was eliminated: Airliners will eventually have a single pilot and a dog on the flight deck. The pilot's job will be to take over if things go wrong. The dog's job will be to bite the pilot if he touches anything.
It would appear as though that joke will become reality.
But without the dog.
7 comments:
yeh, a single pilot on an Airbus' windowless flight deck... what could go wrong...
Doesn't sound like a good idea to me. Maybe the bean counters think halving the crew cost is more than the cost of the occasional hull loss.
Take a look at avherald.com ... there are "crew incapacitated" reports about once a week. Then there's the issue of an automated flight deck and dealing with the *rest* of the issues, which are usually equipment failures.
Sarah is correct.
I currently fly a modern jet airliner. We are getting an upgrade of our FMS (Flight Management System) computers on our fleet of airplanes. The "upgrade" is to the FMS from a Boeing 747-8, the newest 747 out there, just a few years old. The processing speed of the new computers is equivalent to a 286 computer from 1992. The internal memory is measured in megabytes, not gigabytes. The reason is the insanely slow and costly certification procedures required by the FAA. If you extrapolate the time of certification for the "new" FMS to the certification of pilot-less airplanes I think you would reach infinity.
If you then have only 1 crash, and they will crash, it will grind it to a halt. I don't see it happening in my lifetime.
I just hope the pilot doesn't have the fish dinner.
People in the media like to talk about AI. "Skynet became self-aware..." But AI isn't the big thing. Autonomy is. Autonomous machines. Driverless cars. Workerless factories. Aircraft without human pilots. The systems ARE good enough, and dependable enough, and the right redundancy and failure management CAN be built in.
It's a design challenge - good enough never is, but that doesn't mean we can't build systems that work. We just (usually) can't build them on the bid budget.
And so with the ground stations able to take over, if the pilot becomes unable to perform his duties, the easy question is how long until someone decides to take an airliner down because they can. Do we have a drone operator who has used his weapons for his own choice of target? How would we know?
Post a Comment