Droning on...

[ehenders]

Hello all,

I haven't posted much before at all, but I've lurked a lot. Anyhoo, my interests and training are in aircraft, and I'd like to learn more about unmanned aircraft. I was wondering something earlier today, and perhaps there is a technical definition which could answer this or an industry logic, but here's the gist of it:

What is a Drone aircraft? Is it simply an airframe and engine with no wetware controllers aboard the thing itself? Does it need any further levels of autonomy? If you took a highly capable robot, put him in the seat of a cessna and said "go fly around a bit" would that count as a drone?

Probably a good one for the military types, I would think, as they are pioneering drone R&D.

Thanks!
TRH

[CaptainChewbacca]

What is a Drone aircraft? Is it simply an airframe and engine with no wetware controllers aboard the thing itself? Does it need any further levels of autonomy? If you took a highly capable robot, put him in the seat of a cessna and said "go fly around a bit" would that count as a drone?

AHEM

An unmanned aerial vehicle (UAV; also known as a remotely piloted vehicle or RPV, or Unmanned Aircraft System (UAS)) is an aircraft that flies without a human crew on board the aircraft. Their largest uses are in military applications. To distinguish UAVs from missiles, a UAV is defined as a reusable, uncrewed vehicle capable of controlled, sustained, level flight and powered by a jet or reciprocating engine. Therefore, cruise missiles are not considered UAVs, because, like many other guided missiles, the vehicle itself is a weapon that is not reused, even though it is also unmanned and in some cases remotely guided.

My guess would be if the robot is not an integrated part of the drone, then you have a robot pilot flying a manned aerial vehicle. If the robot is hard-wired in as a guidance system, then its still a drone.

[Broomstick]

I haven't posted much before at all, but I've lurked a lot.

A wise choice.

What is a Drone aircraft? Is it simply an airframe and engine with no wetware controllers aboard the thing itself?

Yep, pretty much that's it.

Does it need any further levels of autonomy?

Nope - although that is becoming more common.

If you took a highly capable robot, put him in the seat of a cessna and said "go fly around a bit" would that count as a drone?

Yep.

The FAA's preferred terms these days are UAV or unmanned aerial vehicle or UAS, unmanned aircraft system, which would encompass any group based systems such as a remote pilot or guidance system.

I'm sure the military has a basket of terms and acronyms that go into much more detail than that, but I'm a civilian aviator, not military, so I won't presume to speak for them.

[ehenders]

Interesting. By that logic then these are drones too. So there's no defined difference between a drone and an unmanned aerial vehicle? Do satellites and other unmanned spacecraft count as drones then?

TRH

[UnderAGreySky]

Officially, they need to be above a certain weight limit to be classified as a UAV, at least in the UK. The moment you do that, you've to sign a lot of paperwork saying you won't be selling it yada yada. I have no idea what the limit is, but I have signed such a document for the very same reasons.

[CaptainChewbacca]

Interesting. By that logic then these are drones too. So there's no defined difference between a drone and an unmanned aerial vehicle? Do satellites and other unmanned spacecraft count as drones then?

TRH

Model airplanes are indeed 'drones', but satellites aren't capable of controlled, level, atmospheric flight, so no.

[FSTargetDrone]

"Drone" seems to have a connotation of "dumb," quite different from some UAVs that are capable of autonomous operation. "Drone" can, of course, refer to any remotely-controlled vehicle, be it air, ground or water-based. The US Air Force converts superfluous fighter aircraft into target drones. However, many of these retain the capability to be flown by humans for remote operator training purposes (if the remote operator has a problem, or the link is lost between the ground and the aircraft, the pilot can take over and land the plane), before they are used as targets.

[eion]

There's also the fact that some modern airliners are capable of automated flight from takeoff straight through to landing, which just further blurs the line between UAVs and Manned aircraft.

[Broomstick]

Those airliners aren't yet considered trustworthy enough to fly without on-board human supervision, and operation of the autopilots on them is a bit involved - the pilots I know who fly them would dispute calling them "drones" or "UAV's".

The FAA is currently reviewing the whole unmanned flight thing, from RC models on up. The definitions and regulations may undergo some changes in the next few years.

[eion]

My point was that all too often people get too caught up in semantics. From a technological perspective, the difference between most UAVs and a fly by wire manned-aircraft is the length of the extension cord.

That certainly doesn't mean it is necessarily any easier to fly a UAV vs. a manned aircraft, and often times it can be harder, but the advantage of airliners that can be flown by computer or remotely are quite significant.

[Broomstick]

Yes, among them the fact that if the airplane crashes you won't lose a highly trained pilot, just the "self-loading cargo".

[eion]

Also the fact that you'll have fewwer crashes since your pilots will be either fully rested or not require rest, and many others.

Everythings a trade off.

[Sarevok]

I live near a major airport. I would hate to see a computer controlled Boeing fly into my apartment because of a software glitch. The technology is sound on paper but has a LOT to prove in real world. No one has operated a large aircraft regularly with no humans in the cockpit. The current system of human crew and sophisticated automation works. No need to risk changing it until robotic technology develops a lot further. We don't trust autonomous robots with guns yet. But we would give them right to fly a plane over a city full of people ?

[Shroom Man 777]

Not to mention, with the pilots on board, if there's anything wrong in the plane they might be in a position to do something about it, or at least they would be able to better assess for whatever's wrong by being on the aircraft itself (and not sitting on the ground, interacting with the plane only by remote controls and cameras). From a non-technical standpoint, if I were a customer/passenger it would kind of suck if I rode on the plane where there's the slight risk of dying in an accident while the company's pilot isn't even anywhere near the bird but safe on the ground. From a different (emotional?) standpoint, that's kind of discomfiting, since they (the company) are asking you to ride a plane that they (the company) won't even let their own pilots ride on. Where's the assurance in that.

[eion]

I live near a major airport. I would hate to see a computer controlled Boeing fly into my apartment because of a software glitch. The technology is sound on paper but has a LOT to prove in real world. No one has operated a large aircraft regularly with no humans in the cockpit. The current system of human crew and sophisticated automation works. No need to risk changing it until robotic technology develops a lot further. We don't trust autonomous robots with guns yet. But we would give them right to fly a plane over a city full of people ?

I think you would be surprised how much of your time onboard an airliner you are under the care of the autopilot.

Are they ready for primetime? No, but if airline pilots had to manually fly the plane the entire time we'd have a lot more accidents than we do now.

And of course no one's flown a passenger plane full of people across the country. It'll start with long haul cargo flights, probably allowing the crew to rest while a remote crew supervises the autopilot. Then we'll see over-sea cargo flights with no onboard crew. Then eventually fully automated cargo flights. AND then the whole thing will happen again with passenger airliners.

The first steps already been taken by replacing manned observation and CAS planes with UAVs in most scenarios.

[Sarevok]

I think it would be more prudent for civilian sector to take a wait and see approach. Militaries are not rushing to build drone air superiority fighters despite being a ideal job for a machine. There has to be some good reason why not. Perhaps some fundamental short comings in computer science is to blame. It could be something like image recognition. Computer vision is still in its infancy. I would not trust any autonomous uav which relies on cameras alone to fly reliably. Yet at the same time a good pair of eyes is vital for flying. There is also the fact that computer software is still terrible at decision making. How would a computer controlled plane deal with unexpected component failures while in flight ? You cant program all possible scenarios. For instance the famous incident where a pilot had to use differential thrust from engines alone to turn the plane after hydrolics failed.

[Sarevok]

Edit :

Put simply I think following core technologies have to develop before uavs can take over normal civilian flights.

- computer vision equal to humans. Normally even best algorithoms struggle to beat captcha. And we expect machines to see where they are going in the air ?

- ability to react correctly to unexpected incidents not forseen while the software was created. To replace human pilots the software needs the same flexibility.

- and the ability to make a computer with above capabilities cheaper than hiring a pilot. Even after the tech becomes available the software may require computer equipment not designed or certified to fit into existing planes. And the software itself is going to be very expensive as well. Its going to take a while before airlines consider it as a cheaper alternative to paying pilots to fly planes for them.

[LionElJonson]

Actually, computer vision wouldn't be that important for an autonomous drone aircraft; if it can reliably identify the ground, other aircraft in the air, and runways, it would do just fine. Pilots can't see all that much, and in order to fly the big planes, you need to be able to fly your aircraft without being able to see anything outside the cockpit anyway.

[Sarevok]

How would you identify objects without visual imagery ? Now planes can fly and land at night but they still got so many lights turned on both on aircraft themselves and airport. I imagine while instrumentation can get you near something for safe flying you really need to see where you are going.

[Broomstick]

I live near a major airport. I would hate to see a computer controlled Boeing fly into my apartment because of a software glitch. The technology is sound on paper but has a LOT to prove in real world. No one has operated a large aircraft regularly with no humans in the cockpit. The current system of human crew and sophisticated automation works. No need to risk changing it until robotic technology develops a lot further. We don't trust autonomous robots with guns yet. But we would give them right to fly a plane over a city full of people ?

I think you would be surprised how much of your time onboard an airliner you are under the care of the autopilot.

Yes. An autopilot monitored by a human being who can de-activate it when it malfunctions. Which they do, from time to time.

Computers are best for routine situations. Humans are best for the unforeseen and several types of emergencies. I'm not in a hurry to see that partnership end.

[LionElJonson]

How would you identify objects without visual imagery ? Now planes can fly and land at night but they still got so many lights turned on both on aircraft themselves and airport. I imagine while instrumentation can get you near something for safe flying you really need to see where you are going.

Radar, radio communications, and air traffic control towers, among other things. The only time an IFR (Instrument Flight Rules, as opposed to VFR, Visual Flight Rules) pilot really needs to be able to see is to land. Being able to see is helpful (it makes it a bit more apparent that you're about to fly into the ground), but not really neccessary. 747s have those tiny cockpit windows for a reason, you know.

If you really want to know more, I can crack open the textbook I've got on the subject sitting a few feet away from me; the theory behind IFR flight is what I'm covering this semester for my Aviation classes.

And yeah, humans are good at emergencies, but if you watch Air Crash Investigation, you'll tend to see that the pilots tend to go over the checklist to fix whatever problem they've got a few times. I think the main benefit might well be that human pilots can figure out when to go off the script.

[Sea Skimmer]

Good enough radar will work for that. The latest synthetic aperture array radar has enough resolution to pick out wheel ruts and even wheel marks on wet grass under some circumstances. Detecting other planes and runways would be easy, and unlike visual light it will work in bad weather; though heavy rain will degrade some radar wavelengths. Major airports as well as aircraft carriers also already have instrumented landing systems that can 'capture' a plane and automatically guide it down. Work is going into making that work purely off GPS, I believe the US military already has a GPS landing aid for small UAVs.

The military is also finalizing the design of a very small MMW radar set (couple pounds) it intends to install in drones as small as the RQ-7 Shadow which weighs less then 400lb in total. The radar only has a range of a mile or two, but it has enough resolution to pick out power lines and birds at that distance. Much more could be done on even a small airliner, and we may see the weather radars on current airliners become a lot more capable in the future as AESA will make this easier to do.

Problem is the software just gets more and more complicated, and harder and harder to validate as completely error free. Frankly I see little future for unpiloted commercial airliners, certainly none anytime soon. It just doesn’t make much sense (you don't gain normal UAV advantages like longer flight endurance), and even if the planes do become capable of fully unmanned operation that’s just going to be a backup for the pilot. After all any remote sensing technology you can use to control a UAV could also be used to give better situation awareness to a human pilot. UAV cargo planes are a lot more viable since the cargo doesn’t complain and sue over having its life left in the hand of a machine.

As for now, UAVs are basically a hazard to everything in the air, and only the Global Hawk is allowed to fly in civilian airspace outsize of a war zone. This in turn is only because at 65,000 feet Global Hawk flies much higher then any civilian airliner. So the collision risk is nil, and takeoff and landings take place in military controlled airspace. That allows the ground controllers to simply steer away all the other aircraft while the Global Hawk lands like a dumb happy cow.

[eion]

I live near a major airport. I would hate to see a computer controlled Boeing fly into my apartment because of a software glitch. The technology is sound on paper but has a LOT to prove in real world. No one has operated a large aircraft regularly with no humans in the cockpit. The current system of human crew and sophisticated automation works. No need to risk changing it until robotic technology develops a lot further. We don't trust autonomous robots with guns yet. But we would give them right to fly a plane over a city full of people ?

I think you would be surprised how much of your time onboard an airliner you are under the care of the autopilot.

Yes. An autopilot monitored by a human being who can de-activate it when it malfunctions. Which they do, from time to time.

As an example, the Space Shuttle uses 5 onboard computers. 4 redundant computers and one independent computer that was created without the involvement of the first design team. Similar safety precautions can easily be observed for commercial aircraft.

Computers are best for routine situations. Humans are best for the unforeseen and several types of emergencies. I'm not in a hurry to see that partnership end.

Nor am I at the moment, but it will happen when it's ready.

Computers are also able to react much faster than humans, and monitor far more information than humans. I can easily foresee scenarios where a computer would outreach a human pilot.

The only time an IFR (Instrument Flight Rules, as opposed to VFR, Visual Flight Rules) pilot really needs to be able to see is to land.

If the airport is equipped for CAT IIIc landings and the aircraft has a fail-operational autopilot (meaning that even if the autopilot experiences a failure below alert height it can still complete the landing without human assistance) then the pilots don't even need to see to do that.

[Broomstick]

I think the main benefit might well be that human pilots can figure out when to go off the script.

This is the key - when things really go off the rails and there's no script humans perform better than machines. Which is not to say they're always successful (they aren't) but they can improvise, which machines can not (at least, not with today's tech).

After such an incident, what the humans improvise can then be later studied and incorporated into flight knowledge, both for humans and machines. The two situations where that happened that I can recall offhand were the Airtransit flight that ran out of fuel over the Atlantic where the pilots on board were heard on the CVR referring to the successful unpowered landing in the Gimli Glider incident, and the the DHL cargo plane hit by a missile at the Baghdad airport that, if I recall correctly, utilized knowledge regarding controlling an airplane with differential thrust after hydraulic failure gathered from the Sioux City, Iowa accident. In the latter case, I believe some of that was even programed into the airplane systems.

In any case, although I fully approve of anything that makes the pilots' jobs easier, I still think there's a role for people in the cockpit, and will be for some time. Much to the chagrin of the bean counters who love to dispense with human beings and buy shiny machines to do the work.

[Sea Skimmer]

After such an incident, what the humans improvise can then be later studied and incorporated into flight knowledge, both for humans and machines. The two situations where that happened that I can recall offhand were the Airtransit flight that ran out of fuel over the Atlantic where the pilots on board were heard on the CVR referring to the successful unpowered landing in the Gimli Glider incident, and the the DHL cargo plane hit by a missile at the Baghdad airport that, if I recall correctly, utilized knowledge regarding controlling an airplane with differential thrust after hydraulic failure gathered from the Sioux City, Iowa accident. In the latter case, I believe some of that was even programed into the airplane systems.

I dunno about anything being programed in, its possible, but they knew they had to drop the landing gear to help slow down the plane while using a lot of engine thrust because of the Sioux City crash. Frankly they also just got damn lucky, that on top of it all, that the gear didn't collapse after they ran off the runway and they didn't strike any of the numerous UXOs all over the airport. It took something like a year to finish sweeping the place.