14

enter image description here
(Source)

The Dubai Drone is an Ehang passenger drone planned to start commercial services in July 2017.

If there is a failure, can such an electric multi-copter auto-rotate a landing?

I'm wondering, if an electric motor would even allow auto-rotation, and if the autopilot could handle it? Are there regulations for devices like that?

mike
  • 2,160
  • 19
  • 37
  • 5
    It looks like you could attach a parachute to that, at least. – JAB Apr 24 '17 at 14:37
  • It is presumably possible to build a "drone" rotorcraft with an autopilot capable of flying an auto rotation of a suitable rotor system (though the make the best of a bad situation logic in choosing a forced landing site in unsuitable terrain may be challenging) however the rotor systems of that particular aircraft is not suitable. You'd use independent power systems and/or a ballistic parachute as JAB mentioned. – Chris Stratton Apr 24 '17 at 18:59
  • 1
    Why are people referring to these aircraft as "drones"? A drone, by definition, is robotic and lacking intelligence for rational self-interest (so are generally autonomous for our benefit or are remotely piloted) - it seems people are now generally referring to all quadcopter or multi-rotor aircraft as "drones". – Dai Apr 25 '17 at 08:18
  • 2
    This device is a drone, because it is remotely piloted. – mike Apr 25 '17 at 08:21

4 Answers4

18

That drone has fixed-pitch rotors, and that pitch is optimized for thrust, not for autorotation. In the absence of power, those rotors won't autorotate. They will stop rotating 'in the right way' and then start windmilling in the opposite sense. That windmilling will cause drag and some deceleration of the fall, but not of the same magnitude as an autorotation, that is a different condition.

xxavier
  • 11,071
  • 3
  • 29
  • 72
  • So it will drop quickly? – mike Apr 24 '17 at 07:20
  • 3
    @mike And out of control. The only control of the rotors in that type of drone is by modulating the power of the motors, since the rotors themselves are fixed-pitch and not fully steerable. In the absence of power, there will be little control, if any... – xxavier Apr 24 '17 at 07:28
  • 1
    Also, the inertia of the small rotors looks too small to store enough energy for deceleration. Even if id did have variable pitch props, autorotation will only reduce the sink speed a little. – Peter Kämpf Apr 24 '17 at 07:29
  • @Peter Kämpf I'm not sure... The total disk area is relatively large... It's true that the energy of the rotors is very small, making a final landing flare impossible, but the sink speed will be considerably reduced... (in true autorotation, achieved with the right pitch setting...) – xxavier Apr 24 '17 at 07:40
  • Are there any regulations that guide the development of drones like that? What kind of failure would result in a situation, where autorotation would save a passenger? – mike Apr 24 '17 at 07:41
  • @mike if one motor were to fail, then it could probably execute a powered landing quite safely, though I wouldn't want to do any more than that - it would likely be unstable and underpowered. In this way, the passenger would be saved, but it's not autorotation. – Baldrickk Apr 24 '17 at 12:40
  • @Baldrickk I know, that's why I'm asking about autorotation, because my guess is, that there are quite a few scenarios, that will lead to loss of all rotors, which turns this into the more interesting case. While multicopters are hailed as safer, it turns out they aren't, because they don't autorotate. Imagine living in a city, where several 100 of these things fly and could just drop out of the sky. – mike Apr 24 '17 at 14:45
  • 1
    The way you would make something like this safe would be to have sufficiently independent power systems for each motor that they wouldn't all go down at once. – Chris Stratton Apr 24 '17 at 18:55
14

If a single motor fails, it's not unsafe (supposedly).

The 184 is technically an X-8 multi-rotor, meaning that it has four points of thrust, each consisting of two motor-propellers, coaxially aligned—one “tractor” propeller and one “pusher.” This means that if any one of the motors dies, or a propeller disintegrates, the aircraft won’t flip and crash to the ground, although it would need to land rather quickly. The concept has never been tried in an operational passenger service, however, and some feel that without further safety enhancements the drone taxi idea is a disaster waiting to happen.

— airspacemag.com

The article also discusses how Dubai's hot climate would limit the performance.

As for regulations, so far I can't dig up any, but in the US taxi drones are not allowed.

FAA regulations prevented it from being legally tested in the US. In Dubai, however, the Civil Aviation Authority partnered with EHang in testing the device.

— rt.com

Community
  • 1
  • 1
    I agree, though it's hard to find tests that prove it, if 1 of 8 props fail, you could still execute a powered landing. It might be a bit rough, and you really might not have a large range of "where" you were going to land, but you should be able to hit the "next flat spot". Of course all bets are off if the passenger starts freaking out.To correct for the missing prop the passenger would need to sit very still. – coteyr Apr 24 '17 at 13:57
  • Also it's important to remember "down is easy". 100% of the remaining power and "effort" can go in to simply slowing down the "drop" and making sure you "land" slow enough. – coteyr Apr 24 '17 at 13:59
  • 2
    @coteyr: Actually, "100% of the remaining power" would generally be a bad idea, since you do want the thrust to be approximately symmetrical or the thing will flip over. If you lose one rotor out of eight, you'll probably need to shut down another one (or, more likely, slow down both of the two) on the opposite side. – Ilmari Karonen Apr 24 '17 at 17:24
  • I didn't mean that you turn the props on 100% I mean that you don't need "save" power for going 100 miles you just have to go down. – coteyr Apr 24 '17 at 21:07
  • @IlmariKaronen: Assuming the rotors are all mounted with the diagonals intersecting in the vertical axis through the centre of gravity, even if three rotors on the same diagonal fail, you would still have one rotor left to prevent the thing from rolling around the other diagonal and four rotors for slowing the fall (and stabilizing over the other diagonal), I believe. – Jan Nash Apr 24 '17 at 22:01
  • Although, if you're over very inclined ground, you might still be pretty screwed when having lost three (or even 5!) rotors, since you might come down rather quick and could end up like a rock rolling down the incline... I guess a parachute might be by far the best option there. – Jan Nash Apr 24 '17 at 22:04
11

Simple answer. No.

A drone with fixed wing propellers such as the one you describe can not autorotate. For autorotation to work the pitch of the blades needs to be variable.

At the instant of engine failure, the main rotor blades are producing lift and thrust from their angle of attack and velocity. By immediately lowering collective pitch, which must be done in case of an engine failure, the pilot reduces lift and drag and the helicopter begins an immediate descent, producing an upward flow of air through the rotor system. This upward flow of air through the rotor provides sufficient thrust to maintain rotor rotational speed throughout the descent. Since the tail rotor is driven by the main rotor transmission during autorotation, heading control is maintained as in normal flight. - https://en.wikipedia.org/wiki/Autorotation

A drone or multicopter with 6 or more propellers can continue flying as long as the thrust provided by the remaining propellers and engines is sufficient to keep it from falling.

The aicraft may however begin to rotate around its own axis. In normal flight there are an equal number of clock wise and counter clockwise rotating propellers. If a propeller/motor stops spinning the torque of these engines will no longer cancel each other out so in order to not descend the helicopter may be forced to accept a rotation around the Z axis. This same mechanism is used for turning a multirotor, lower the RPM of the propellers spinning in one direction and you will turn in the other.

I fly RC drones myself and have tested how my hexacopter behaves during a motor failure. The conclusion I came to is that if it was close to its carrying capacity it would rotate since it needs all the power it has and can't worry about rotation. But if it only carried a light load it woulden't rotate as it has power left over to compensate for it.

There are however a few attempts at making drones with variable pitched propellers Variable pitch drone

This however negates the pure simplicity that a drone provides as one of it's main advantages is that there are only as many moving parts as there are propellers.

In my opinion and others (http://diydrones.com/forum/topics/why-are-we-not-seeing-more-variable-pitch-quadcopters) the only way to provide a reliable drone is to have 6 or more propellers and lots of power to spare.

Community
  • 1
JensB
  • 211
  • 2
  • 5
  • 1
    "A drone or multicopter with 6 or more propellers" You mean 4. One could design a drone that had 4 rotors that could handle a motor out condition. – Sam Apr 24 '17 at 14:12
  • 1
    @Sam true, although I have yet to see one. Also, with only 4 props loosing one would mean loosing a substantial amount of the total thrust which would mean that you would have to create something very overpowered or be forced into a descent upon failure. In addition the CG of the craft would be much more crucial and even a small mistake or misloaded cargo would cause a crash. – JensB Apr 24 '17 at 14:27
  • I think an odd number of rotors is the best. Like for multiblades propellers or fans, if you lose one blade, desiquilibrium is less dramatic with an odd number of blades than an even one. For instance, heptarotors are safer than octorotors, since they can both lose a maximum of 4 rotors. 4/7 is better than 4/8 ratio. – user721108 Apr 24 '17 at 15:14
  • 1
    @qqjkztd What? "heptarotors are safer than octorotors" they may be more "safety efficient" but are not generally safer. JensB is correct the torque of rotation is a design problem with odd rotor numbers. Also when you get beyond 1 lost rotor, you have to account for where and how the symmetry works out. – Sam Apr 24 '17 at 16:09
  • @Sam I'm talking about contrarotative propellers driven by one engine, like seen in related question about Dubai drone. There is no yaw stability issue when losing one or more engines, since each brushless engine dirves two propellers (1 fixed axle - stator 1 prop - rotor 1prop) – user721108 Apr 24 '17 at 17:10
  • @qqjkztd then you cant turn as the multicopter uses the torque of the engines to turn. – JensB Apr 24 '17 at 17:23
  • Another intresting point in this is that the multicopter falls like a rock if any of the electric systems fail (which in my hobby experience is waaayyy more common than broken engines/props) where as a normal helicopter can keep going or at least land in most cases. Personally I would rather travel in a regular helicopter any day of the week. – JensB Apr 24 '17 at 17:26
  • @Sam, how do you handle engine out with just 4 engines? You have 3 left, but because the CoG is between the two, the third mustn't be producing any lift, which it can't do without variable-pitch propeller. And if you cut it, you lose roll control on that axis. – Jan Hudec Apr 24 '17 at 17:32
  • @qqjkztd, I understand it has 8 engines, each driving its own propeller. If it was just 4 engines, it wouldn't be controllable with one engine out. – Jan Hudec Apr 24 '17 at 17:33
  • @JensB you're right, the picture shows one engine for each propeller. In order to have yaw control on a multicopter with single engines driving two contrarotative propellers, other types of yaw control would be necessary, adding weight and complexity. (small ailerons under each engine may be light but draggy in horizontal flight, tilting engines is heavy...) – user721108 Apr 24 '17 at 17:51
  • @JanHudec , I said one could be designed that way. It wouldn't work in the configuration you are used to. You could design one with the engines in a pyramid, for example, or with 3 arms and a center engine. – Sam Apr 24 '17 at 18:00
  • @Sam, centre engine won't help unless it alone can keep the thing flying. Pyramid would, since you can tilt to move the CoG inward, but it would need tilting engines. – Jan Hudec Apr 24 '17 at 18:09
  • @JanHudec I didn't say it wasn't a trivial case, I just said it could be done. – Sam Apr 24 '17 at 18:57
  • @JensB have you signed up to the Drones proposal on Area51? https://area51.stackexchange.com/proposals/107036/drones – Rory Alsop Dec 07 '17 at 07:44
7

I am a career helicopter pilot who enjoys watching the development of drone technology. I can tell you for certain that this aircraft is incapable of autorotation. A parachute would be it's best option for pure simplicity. This would not guarantee a safe landing, just avoid a 'splat'. It is very apparent to me that the number one challenge of the drone industry as it relates to increased payload (passengers, cargo, luggage) is power. Drone advancement is highly reliant on battery advancement. If you have a lightweight, high powered battery system, then the sky's the limit for drones, safety and otherwise. Short of that, I'm still waiting for the big breakthrough. For now, I'd just like to see y'all drone drivers get an anti-collision light. Oh wait...more power is going to be needed for that :)

user27970
  • 81
  • 1
  • 1