Go Back  Airline Pilot Central Forums > Pilot Lounge > Aviation Technology
The Future Of Artificial Intelligence >

The Future Of Artificial Intelligence

Search
Notices
Aviation Technology New, advanced, and future aviation technology discussion

The Future Of Artificial Intelligence

Thread Tools
 
Search this Thread
 
Old 11-04-2018, 01:59 AM
  #141  
In a land of unicorns
 
Joined APC: Apr 2014
Position: Whale FO
Posts: 6,473
Default

Originally Posted by Name User View Post
Remember when FedEx converted their DC-10's and removed the flight engineer? And that was years ago.
It took them 26 years to do. (DC10 first flight in 1970 and MD10 in 1996).
Or 20 years, the R&D for DC10 replacement was started in 1976.

It takes 10+ years to develop a traditional airliner. And there's nothing in public pipeline with 1 pilot. Pretty hard to see a 1 pilot plane in 121 ops in 10 years. Maybe 20, and that's for cargo only.
dera is offline  
Old 11-04-2018, 02:46 AM
  #142  
Gets Weekends Off
 
captjns's Avatar
 
Joined APC: Feb 2006
Position: B-737NG preferably in first class with a glass of champagne and caviar
Posts: 5,914
Default

Originally Posted by dera View Post
It took them 26 years to do. (DC10 first flight in 1970 and MD10 in 1996).
Or 20 years, the R&D for DC10 replacement was started in 1976.

It takes 10+ years to develop a traditional airliner. And there's nothing in public pipeline with 1 pilot. Pretty hard to see a 1 pilot plane in 121 ops in 10 years. Maybe 20, and that's for cargo only.
At a first group meeting, attended by UAL, JAL, AA, and BA to name a few, held in January 1990, a 23-page questionnaire was distributed to the airlines, asking what each wanted in the design. By March of 1990, preliminary designs, for the B777, were started. First Flight... June, 1994, and entered into service, by UAL, a 121 operator, June of 1995.
captjns is offline  
Old 11-04-2018, 07:32 AM
  #143  
Gets Weekends Off
 
Adlerdriver's Avatar
 
Joined APC: Jul 2007
Position: 767 Captain
Posts: 3,988
Default

Originally Posted by dera View Post
It took them 26 years to do. (DC10 first flight in 1970 and MD10 in 1996).
Interesting math and logic. You make it sound like the second the first DC-10 lifted off in 1970, development of the MD-10 began. 26 years later it was finished. Especially since FedEx didn’t start operations or flying aircraft until 1973 and didn’t get a DC-10 until 1980.
Isn’t that like saying it took 91 years to develop the 777? (Wright flyer 1903.....777 1994)
Adlerdriver is offline  
Old 11-04-2018, 10:47 AM
  #144  
Gets Weekends Off
 
Adlerdriver's Avatar
 
Joined APC: Jul 2007
Position: 767 Captain
Posts: 3,988
Default

Originally Posted by Name User View Post
You don't need a new airframe to integrate an AI like autopilot.

Just think of a typical takeoff. After rotation we must raise the gear and flaps. Why? Why isn't that automated? Or on landing, it can easily be automated as well. "Drop gear at xxxx altitude". Requested airspeed is xxx, place flaps at x. Etc.

ATC tells us what altitude to climb to, they tell us the route. There is no reason you couldn't integrate a higher level autopilot to completely take over operating the aircraft. The pilot would be there to deviate around weather and make command decisions and taxi.

In an emergency, say an engine out, the advanced autopilot would know the runway and location and execute perfect E/O procedures. It could also simultaneously detect a fire and put it out. All while you watch.

Have you used the newer drone software available today? You tell it where you want to go and it handles the rest. I envision that will be what aviation is like in the next generation. You are there to monitor.

I personally don't see aircraft operated from the ground at this point. But a reduction in pilots required? Sure.
Once again, we have someone on here who is confusing the technological ability to automate existing aircraft (which we obviously have) with the practical ability to make it happen in the current industry/regulatory/ATC environment.
Reducing pilots from two to one really means fully automated aircraft to account for the possibility the one pilot becomes incapacitated. So that basically means aircraft operated from the ground (which you just said you don't see).

How do you automate out potential human errors in procedures or mis-heard clearances that are currently caught by the other pilot. Until we go to ATC that doesn't rely on spoken VHF comm, it's a huge reduction in safety to expect one pilot to operate in our busiest airports.

Those are just a couple of major roadblocks in a very long list. Meanwhile, Boeing and Airbus are pumping out brand new 777F, 787s, 737s, A530s, etc. with expected lifespans of decades that still need two pilots. Those companies aren't going to spend significant dollars to do much of anything until airlines and cargo companies are ready to buy. That's certainly not going to happen until there are major changes to the industry infrastructure. And all that just to take one pilot out of the aircraft and replace him with one on the ground? Frankly, we don't get paid that much. I seriously doubt anyone posting here will see a practical, production or modified single pilot 121 aircraft in revenue service before they are required to retire (at 65).
Adlerdriver is offline  
Old 11-04-2018, 10:48 AM
  #145  
Gets Weekends Off
 
Joined APC: Mar 2014
Posts: 3,097
Default

Originally Posted by dera View Post
It took them 26 years to do. (DC10 first flight in 1970 and MD10 in 1996).
Or 20 years, the R&D for DC10 replacement was started in 1976.

It takes 10+ years to develop a traditional airliner. And there's nothing in public pipeline with 1 pilot. Pretty hard to see a 1 pilot plane in 121 ops in 10 years. Maybe 20, and that's for cargo only.
With current technology we don't need a single pilot designed airplane. Did you click the link and watch the video?
Name User is online now  
Old 11-04-2018, 10:56 AM
  #146  
Gets Weekends Off
 
Joined APC: Mar 2014
Posts: 3,097
Default

Originally Posted by Adlerdriver View Post
Once again, we have someone on here who is confusing the technological ability to automate existing aircraft (which we obviously have) with the practical ability to make it happen in the current industry/regulatory/ATC environment.
Reducing pilots from two to one really means fully automated aircraft to account for the possibility the one pilot becomes incapacitated. So that basically means aircraft operated from the ground (which you just said you don't see).

How do you automate out potential human errors in procedures or mis-heard clearances that are currently caught by the other pilot. Until we go to ATC that doesn't rely on spoken VHF comm, it's a huge reduction in safety to expect one pilot to operate in our busiest airports.

Those are just a couple of major roadblocks in a very long list. Meanwhile, Boeing and Airbus are pumping out brand new 777F, 787s, 737s, A530s, etc. with expected lifespans of decades that still need two pilots. Those companies aren't going to spend significant dollars to do much of anything until airlines and cargo companies are ready to buy. That's certainly not going to happen until there are major changes to the industry infrastructure. And all that just to take one pilot out of the aircraft and replace him with one on the ground? Frankly, we don't get paid that much. I seriously doubt anyone posting here will see a practical, production or modified single pilot 121 aircraft in revenue service before they are required to retire (at 65).
With automation flying the airplane it will free the pilot up to strictly monitor everything.

CPDL is already operating domestically, being trialed overnight in Indy center IIRC. This will already significantly reduce hear/readback errors. It's not a stretch of our imagination to simply route that clearance directly into the automation.

Pilots in the future will simply tap/click on a flightplan and the aircraft will execute it. Including requests for step climbs and shortcuts. All of this can and will be automated, and it will do it the same way every single time. I still see single pilot ops, I do not see drone airliners. Yet. But I do see drone fleets of FedEx Caravans.
Name User is online now  
Old 11-04-2018, 11:52 AM
  #147  
Gets Weekends Off
 
tomgoodman's Avatar
 
Joined APC: Feb 2006
Position: 767A (Ret)
Posts: 6,248
Default

Regardless of what is technically possible, it costs too much, too soon, and pays back too little, too late.
That violates Beancounter Rule #1: “Never impoverish yourself to enrich your successor.”
tomgoodman is offline  
Old 11-04-2018, 12:15 PM
  #148  
Gets Weekends Off
 
Adlerdriver's Avatar
 
Joined APC: Jul 2007
Position: 767 Captain
Posts: 3,988
Default

Originally Posted by Name User View Post
With automation flying the airplane it will free the pilot up to strictly monitor everything.

CPDL is already operating domestically, being trialed overnight in Indy center IIRC. This will already significantly reduce hear/readback errors. It's not a stretch of our imagination to simply route that clearance directly into the automation.

Pilots in the future will simply tap/click on a flightplan and the aircraft will execute it. Including requests for step climbs and shortcuts. All of this can and will be automated, and it will do it the same way every single time. I still see single pilot ops, I do not see drone airliners. Yet. But I do see drone fleets of FedEx Caravans.
You're still confusing technological possibilities with economic practicalities. Tom G. hit the nail on the head.

CPDLC use by a center ATC facility to provide enroute control is a far, far cry from using it to bring aircraft into EWR, ORD or LAX with 5 mile spacing.

ATC modernization is going to happen because it's going to enhance safety. Because of that, there will be public funds available to improve and expand that system. You can't make the same argument about reducing onboard pilots. The only reason anyone would do it is for purely economic reasons, which potentially benefit only the private companies operating those aircraft. So, the expense required to modify aircraft, ground and air infrastructure, employ ground pilots/monitors, etc. all have to be covered by private funds. Public tax dollars are not going to be allocated to ensure that FedEx or American can save money by reducing pilots on board their aircraft. Once again, you're talking about a huge outlay of R&D dollars and major capital expenditures with the only end benefit being one less pilot. In the big scheme of things, we're not that expensive when you compare us to the overall industry cost of automating current or future aircraft. Let's talk again in 30 years and if I'm wrong, you can say "I told you so", but I'm not.
Adlerdriver is offline  
Old 11-04-2018, 03:46 PM
  #149  
In a land of unicorns
 
Joined APC: Apr 2014
Position: Whale FO
Posts: 6,473
Default

Originally Posted by captjns View Post
At a first group meeting, attended by UAL, JAL, AA, and BA to name a few, held in January 1990, a 23-page questionnaire was distributed to the airlines, asking what each wanted in the design. By March of 1990, preliminary designs, for the B777, were started. First Flight... June, 1994, and entered into service, by UAL, a 121 operator, June of 1995.
Look at 787 and A350, both took close to 10 years. Partly because of new manufacturing methods required, and because of market requirements. Both would play a huge role on development of a single pilot plane.
It's harder today to design a new airplane that someone would actually buy, the market is pretty saturated with existing models for pretty much any segment you can imagine.
dera is offline  
Old 11-04-2018, 04:13 PM
  #150  
Gets Weekends Off
 
Joined APC: Jun 2017
Posts: 659
Default

If and when single pilot or autonomous airliners happen, it will come from small to large. A single piloted airliner must capable of autonomous or extremely reliable ground controlled landings. It will start with small Uber type VTOL autonomous aircraft then get scaled up as the technology evolves and passengers become comfortable with it.
Peacock is offline  
Related Topics
Thread
Thread Starter
Forum
Replies
Last Post
JetJock16
Regional
278
03-10-2017 02:03 PM
par8head
Money Talk
31
12-23-2015 03:03 AM
FloridaGator
Hangar Talk
26
10-02-2008 10:24 AM
flyharm
Mergers and Acquisitions
5
09-11-2008 05:08 PM
maximaman
Regional
31
09-03-2007 05:38 AM

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On



Your Privacy Choices