Jeremy Vine Show - Pilotless Airliners

[evansb]

There is no such thing as a 100% safe manned aircraft either. Otherwise, actuary tables would reflect a significantly reduced risk..

I predict unmanned jet freighters between major terminals by 2027. Unmanned semi-trucks (lorries) are ready to roll in temperate climates on undemanding routes. Unmanned 1-mile long freight trains without cabooses already exist.. Yes, the trains and lorries will operate in a protected, somewhat predictive and controlled environment, but the precedent has been set..unfortunately.

Unmanned passenger airliners? Probably never. Public perception, public relations, liability issues and corporate images just might outweigh the slickness and efficiency of automation.
Surely we all want a "Captain", even on the highly computerized (pre floppy disk era) Starship NCC-1701 Enterprise...don't we?

[Ian W]

Originally Posted by evansb

There is no such thing as a 100% safe manned aircraft either. Otherwise, actuary tables would reflect a significantly reduced risk..

I predict unmanned jet freighters between major terminals by 2027. Unmanned semi-trucks (lorries) are ready to roll in temperate climates on undemanding routes. Unmanned 1-mile long freight trains without cabooses already exist.. Yes, the trains and lorries will operate in a protected, somewhat predictive and controlled environment, but the precedent has been set..unfortunately.

Unmanned passenger airliners? Probably never. Public perception, public relations, liability issues and corporate images just might outweigh the slickness and efficiency of automation.
Surely we all want a "Captain", even on the highly computerized (pre floppy disk era) Starship NCC-1701 Enterprise...don't we?

This is more of an emotional argument, that the SLF require the comfort blanket of a crew at the front, or they will be unwilling to fly. This despite the fact that the crew only perhaps control the lift off and then don't touch the controls until after the aircraft is slowing from its CAT IIIb autoland (used as crews are not capable of sufficient safe accuracy).

The human-on-the-loop systems of today are close to autonomous. The crew are there to take over in the 'otherwise' cases that are expensive to design and certify. But the cost of the crew is increasing and the otherwise cases are being solved in military systems - yes at the cost of attrition in the learning environment - but now solved. All UAS are required to be autonomous for the occasions when the control link fails. In Lost Link operations by definition the recoveries are autonomous.

There is no extra technology or problem with changing normal freight for SLF, just a psychological boundary and for the regulators more certification testing. Validation and verification testing will already have been carried out for the military operations so the extra certification testing should not prove too challenging just long.

It is the psychological boundary that is difficult, especially for those with a vested interest in retaining their front row seat.

[Arfur Dent]

Ian W - what do you mean by "CAT 111B used as crews are not capable of sufficient safe accuracy"??
With your use of half truths and semi facts, you are obviously not a pilot. Do you know how often pilots carry out CAT 3 autolands? Do you know what happens to flow rates at, say LHR, when LVPs are in force? Do you know what facilities are required regarding airborne and ground equipment to carry out LVPs?
When LVPs are not in force and sensitive areas are not required to be protected, almost every pilot will carry out a manual landing flying large parts of the final approach manually as well.
Yes, we have a vested interest as pilots but we do as passengers too. How can you possibly know how many millions of people have been saved by expeditious and timely pilot intervention? After you on the first pilotless aircraft - I won't join you - ever.

[Ian W]

Originally Posted by Arfur Dent

Ian W - what do you mean by "CAT 111B used as crews are not capable of sufficient safe accuracy"??
With your use of half truths and semi facts, you are obviously not a pilot. Do you know how often pilots carry out CAT 3 autolands? Do you know what happens to flow rates at, say LHR, when LVPs are in force? Do you know what facilities are required regarding airborne and ground equipment to carry out LVPs?
When LVPs are not in force and sensitive areas are not required to be protected, almost every pilot will carry out a manual landing flying large parts of the final approach manually as well.
Yes, we have a vested interest as pilots but we do as passengers too. How can you possibly know how many millions of people have been saved by expeditious and timely pilot intervention? After you on the first pilotless aircraft - I won't join you - ever.

Actually, I do know all the limitations of ILS, which is rather too slowly being replaced by GLS and GBAS which is just as accurate (if not more accurate) and does not have issues with multi-path reflections requiring protected zones and extra in-trail separation. Perhaps you should read up about those, as GLS will be fitted to more advanced aircraft and they will not need to have extra separation. Perhaps you will then think about the number of days at Heathrow and other major hubs where the runway acceptance rate is severely reduced by aircraft using ILS when those aircraft using GLS could maintain the acceptance rates. An entire GLS/GBAS system can be installed at an airport for the cost of one annual ILS calibration and does not have the ILS limitations and will provide CAT IIIb accuracy for GLS approaches to all the runway ends within 20KM. GLS aircraft will save both the aircraft operators and airport operators significant amounts of money and disruption recovery exercises.

Not only that but as has been shown in developments in Seattle/Tacoma (SEA), curved instrument approaches can be carried out allowing reductions in flight time and noise nuisance while maintaining runway acceptance rates.

Autonomous operations are already being carried out in military developments and as said in this thread that includes carrier landings and air to air refueling both of which are far more complex than a simple IMC landing on a fixed runway. The potential financial benefits are not lost on the aircraft operators. As is always the case it is the regulators that cannot keep up with the capabilities of new aircraft. Although there are already UAS that are Part 23 certified.

Expect significant changes in the next decade and by 2035 the aviation world will have changed more than it did with the introduction of jet engines.

Addition>
I would have no problem being flown in a UAS. Enjoy your life in the aerospace museums saying "You won't get me up in one of those things."

[Basil]

the crew only perhaps control the lift off and then don't touch the controls until after the aircraft is slowing from its CAT IIIb autoland

In-flight decisions are an ongoing process.
For instance, I have twice been told by ATC to turn around and go back, on one occasion as we penetrated a dense line of Cb when the shortest exposure to hazard was to continue (the pros will understand reduced margins in the turn and degraded Wx radar coverage). The other would have been expensive, hazardous and inconvenient and was clearly due to a misunderstanding about baksheesh. On both occasions we continued.
What would a computer have done?
How will a computer handle a military interception? Climb, descend, turn? Good way to get yourself shot down.

I could go on but really, unless you are a professional aviator, stick to asking questions instead of making statements.

[OldLurker]

In-flight decisions are an ongoing process.
For instance, I have twice been told by ATC to turn around and go back ... What would a computer have done?
How will a computer handle a military interception? Climb, descend, turn?

As is frequently pointed out in this forum, an aircraft's commander is responsible for its safe operation and not only is entitled, but has a duty, not to do what ATC says if the situation merits. An artificial-intelligence 'captain' of an autonomous aircraft would surely have the same authority. An aircraft that can be controlled by ATC wouldn't be an autonomous pilotless aircraft but a piloted UAV with ATC having the command responsibility.
A military interception is a bit different. Then the AI 'captain' would have to be able to understand that it's being intercepted, understand commands from the interceptor and do what it's told. Alternatively the interceptor could be equipped to take control, so that the aircraft again becomes a piloted UAV with the interceptor having command responsibility.

[Ian W]

OldLurker and Basil,
You are both aware that UAS are flying in combat situations continually where dealing with intercepts is a little more common than your experience? Global Hawk's don't get shipped out they fly out to the remote areas they work in and yes through the ITCZ and in bad weather. Just because you don't understand how all these 'continuous decisions' are being made in the UAS world doesn't mean that they are not.

[OldLurker]

Ian W, a bit of knee-jerk there. If you read what I said, and what I replied to, you might have understood that I was talking about autonomous pilotless aircraft rather than remotely-piloted UAS such as the Global Hawk, and about interception of civilian aircraft by military aircraft in normal airspace not war-zone (e.g. in the case of a security concern, or cases like Helios 522), which is never done by UAS!

[Arfur Dent]

Ian W
Still no answer to "What did you mean by CAT 3B used as crew are not capable of sufficient safe accuracy?". What did you mean and what is "sufficient safe accuracy"?
I am aware that the ILS will eventually be replaced but I think most aircraft will have pilots for the next generation or 2 - at least.

[Basil]

You are both aware that UAS are flying in combat situations continually where dealing with intercepts is a little more common than your experience?

Really? With 400 pax on board? In a situation where an ATC handover has failed and a fighter appears in one's 10 o'clock rocking her wings?
Nope, I don't think we are close to having computers handle the sort of situations I've mentioned.

[dr dre]

Expect significant changes in the next decade and by 2035 the aviation world will have changed more than it did with the introduction of jet engines.

Be careful about future predictions of the advance of technology. I bet on the day of the Concordes first flight had I told you that less than 20 would be built and in 30 years time we'd be in subsonic aircraft crossing the Atlantic at the same speed as a 707 with no plans to go any faster I would have been laughed out of the room.