[libranof1987]

While Boeing struggles to fix a flaw with one of their plane's software (that causes it to crash), Airbus is testing pilotless flying! Would more reliance on sensors and software result in more ways things can go wrong if something malfunctions.

According to Airbus, the A350-1000 achieved eight automatic takeoffs over a period of four and a half hours.

"While completing alignment on the runway, waiting for clearance from air traffic control, we engaged the autopilot," Airbus test pilot Captain Yann Beaufils explained in a statement.

"We moved the throttle levers to the takeoff setting and we monitored the aircraft. It started to move and accelerate automatically maintaining the runway center line, at the exact rotation speed as entered in the system."

"The nose of the aircraft began to lift up automatically to take the expected takeoff pitch value and a few seconds later we were airborne."

More hardware + more software = ?

This was achieved via new image recognition technology installed directly on the aircraft, rather than an Instrument Landing System (ILS), which sends radio waves up from the runway, providing pilots with vertical and horizontal guidance.

This, will make the entire sequence (from taxi to landing) completely autonomous.

Airbus says it also plans to trial automatic vision-based taxi and landing sequences later this year.

Scared or nervous about this? Needn't be, because

n 2019, a survey of 22,000 people by US software firm Ansys indicated that 70% of travelers would be prepared to fly in fully autonomous aircraft.

https://edition.cnn.com/travel/artic...ets/index.html

[Jeroen]

I am surprised at the high percentage of people that would be prepared to fly with no pilot!

As it is many people are reluctant when it comes to autonomous driving.

Anyway, yes in principle more automation will make flying more safe. Automation to date has contributed greatly to aviation safety as it stands.

To date, the fast majority of fatal accidents tend to have a high degree of pilot error involved.

Problems such as with the Max design are rare. Although, given what we know about how Boeing and the FAA dealt with this design, a further revisit of design/test/verification rules for autonomous flight might well be in order.

You won’t find me boarding an autonomous plane any time soon!

Jeroen

[RahuKetu]

This may sound little bit irrelevant to this topic but during my stint with "Delhi Metro Rail Corporation" I learnt that there were three modes of operation of trains viz Manual, Semiautomatic & Fully automatic. Trains ran on semi auto mode at that time with train operators on board, I am talking about 2006.

The trial run of India’s first ever Metro train, equipped to run on the ‘Unattended Train Operation’ mode, was ceremonially flagged off by then Union Minister for Urban Development, Sh. M. Venkaiah Naidu and Delhi Chief Minister, Sh. Arvind Kejriwal.

The inaugural train moved out from the Mukundpur depot and travelled up to the Majlis Park Metro station of the Majlis Park – Shiv Vihar Metro corridor of Phase 3 (Pink Line). The new trains, with extremely high levels of automation were due to undergo rigorous trials to ensure that they are ready for smooth operations after the commissioning of the new corridors. Initially, train operators will run the trains, but gradually, driver less operations (on the Unattended Train Operation (UTO) mode) will be possible.
Dubai metro is running on fully automatic mode.

People have started to trust systems without human interference, aviation system however is a different ball game. I know its little bit idiotic on my part to compare planes with trains since latter has more risks & complexities involved. Moreover when automation fails then it is the pilot who controls the aircraft. When automation fails, it fails big time. Lion Air plane crash in 2018 is a testimony to that. I wouldn't personally prefer to fly without a captain in the cockpit whether he is flying or sleeping I don't care.

[Malyaj]

Originally Posted by Jeroen

I am surprised at the high percentage of people that would be prepared to fly with no pilot....

....Anyway, yes in principle more automation will make flying more safe. Automation to date has contributed greatly to aviation safety as it stands.

To date, the fast majority of fatal accidents tend to have a high degree of pilot error involved.

These 70% who are comfortable with pilot-less aircraft are probably just show-offs pretending to be adventurous people !!

I am no pilot but I do believe that automation has reduced accidents caused due to human error, not just in aviation but in all industries. My interest lies in knowing how the role of pilots is changing from being a one who actually flies the aircraft to one who operates a myriad set of computer systems. I know you fly as a hobby GA pilot but do you have any insights on this or from the commercial pilot fraternity (who are invariably the ones who fly with these huge glass cockpits) on what they feel about the change? The ones who have flown in both the eras, do they feel that they are better prepared to handle emergency situations compared to modern pilots?

[Jeroen]

Originally Posted by Malyaj

I am no pilot but I do believe that automation has reduced accidents caused due to human error, not just in aviation but in all industries.

Yes, true, for aviation and other industries

Originally Posted by Malyaj

My interest lies in knowing how the role of pilots is changing from being a one who actually flies the aircraft to one who operates a myriad set of computer systems. I know you fly as a hobby GA pilot but do you have any insights on this or from the commercial pilot fraternity (who are invariably the ones who fly with these huge glass cockpits) on what they feel about the change? The ones who have flown in both the eras, do they feel that they are better prepared to handle emergency situations compared to modern pilots?

I am a GA pilot, but I used to fly pretty advanced planes with glass cockpit and very advanced avianics too:

https://cirrusaircraft.com/innovatio...pective-touch/

These planes are very capable, highly automated and will get you all the way up to 25000 feet where you battle it out with the big fast boys!

In fact you will find that many avionic innovations came to the GA planes before they were introduced into the commercial planes. That is partly because there are less complex system to interact with, but also the certification requirements are less and therefor a lot less costly.

Cirrus has just launched a “safe return” option, a fully autoland system. If the pilot becomes incapacitated, a passenger can hit a button and the aircraft will automatically land at the nearest and safest airport! I don’t see any commercial jets doing that any time soon: Once that button has been pushed it becomes a fully autonomous aircraft. It will look at hazardous terrain, wether, winds aloft, it communicates with ATC etc!! Have a look at this:

https://www.youtube.com/watch?v=PiGkzgfR_c0

GPS was allowed as primary navigation source on GA aircraft long before it made its way to commercial aviation.

But all pilots learn to fly on single engine GA planes, first under visual flight rules only, next typical an instrument rating, various endorsements, ATP certification and ultimately some sort of type rating on a commercial jet or otherwise.

So all pilots are taught the basic of flying an aircraft by themselves. They learn to do the flight planning, flight prepping all by themselves. Only when you have sufficient proficiency in doing it by yourself, does it makes sense to start using automation. At least, you will find that is the consensus among most pilots and instructors.

But with automation you do become more of a system manager. And the problem is that manual flying skills tend to erode. People will look at their screens, even when they should be looking outside. They are questioning the computer, whereas they should have just take manual control.

As it is, many professional pilots will log up hundreds of hours of flight time every month, but within those many hours only a few minutes of hand flying.

If there is one hot topic being debated on any pilot forum these days, it is about this very topic. Young pilots do not have enough experience hand flying. Remarkably, nobody has actually done a proper study on this phenomena I think.

But, of course, with any skill it is use it or loose it.

A here is the thing; humans are not so good at staying attentive during very repetitive task. Such as staying real alert during a cruise across the ocean. Usually not a huge thing, you are up high, so you have time to respond.

But then again, the other thing humans are not particularly good at, is responding on things that startle you. To some extend training will help, but many aviation accident show that even experienced and seasoned crews can get it badly wrong.

This particular accident and detailled research came into the news recently. Have a scan through it, it highlights various aspects on what manufactures and legislators might believe as to how pilots should respond, but the reality is very different on what you should assume.

https://www.onderzoeksraad.nl/nl/med...t_s_dekker.pdf

I am a relative low hours pilot, flying pretty advanced GA planes. So I had all the automation at my disposal. And lets face it, flying a plane for hour after hour during cruise is not very exiting, so you use the auto pilot.

But here is the thing; the auto pilot is also a great tool for lowering pilot workload. So for instance, during descent and approach to an unknown airport it might make a lot of sense to have George, your friendly auto-pilot, fly the plane. That leaves you with time and focus on communicating with ATC, understanding the tactical situation, monitoring etc.

On the other hand, it does rob you of experience.

So in my case, whenever I had the time and fuel I would do it twice. When flying to a new airport I had never finished before, I would use the autopilot on the first landing, take off again and fly the whole approach sequence manual. Especially during poor weather.

Remarkably, passengers on jets would take a dislike to such skill and experience exercise.

Jeroen

[alpha1]

There is a problem I see with full automation in planes.
What is the Fail Safe condition here?

To give an example: if fully automated train system determines itself (using internal diagnostics, voting etc.) that it has malfunctioned - the system will stop the train movement. Same thing for any other land based transport. And people will be safe.

But what happens in air?

[Jeroen]

Originally Posted by alpha1

There is a problem I see with full automation in planes.
What is the Fail Safe condition here?

To give an example: if fully automated train system determines itself (using internal diagnostics, voting etc.) that it has malfunctioned - the system will stop the train movement. Same thing for any other land based transport. And people will be safe.

But what happens in air?

Same problem with the pilots, they do not recognise themselves when they are “malfunctioning”.

Hence the endless string of aviation accidents due to pilot error as one of the leading causes.

Jeren