So the pentacontakaitetracopter...a practical design?

In My Humble Opinion
1

So, as Stranger is no doubt going to soon point out, if an aerospace engineer cooked up this pentacontakaitetracopter, they probably would soon be washing dishes in a soup kitchen.

But is it so bad? The one thing I can think of that is an advantage is :

  1. It’s really really easy to control. You can just divide the 54 rotors into zones that correspond to the 4 props on a quadcopter in code. There’s more complex things you can do to correct for failed rotors and such but the basic quadcopter PID control loop everyone does is enough to get off the ground. I can’t tell how he is getting his speed signal to each rotor controller…it might be wireless…and there is a single point of failure in that he probably just uses a hobbyist joystick and a single arduino or something.

Redundant avionics would be and something you would want to do to make one that was semi reliable.

  1. It’s really cheap. It actually flies, and if you could solve the control reliability, would let you fly for about 10 to 30 minutes depending on how many batteries you loaded on it. The actual rotor system is highly reliable - as you can see, each node is isolated and has it’s own battery and motor controller. Ironically, I suspect this kind of architecture is probably less likely to suddenly fail than a multimillion dollar turbine helicopter. Sure, each cheap chinese made rotor and motor and battery is highly likely to fail at any moment, but the probability that enough of the 54 suddenly fail all at once before you can land is low. You just need a control system able to recognize when nodes have failed and to advise a landing to the pilot. You also need a control system that prevents unrecoverable flight states - I’d want the inputs from the pilot to be only requests to fly a particular direction, and for the control loop to prioritize level, stable flight over all else with no lateral velocity.

How cheap? 54 rotor packages times about $250-$500 of parts for each one. So about $27,000 in parts. You could manufacture these things if you didn’t care about liability for $81,000 each.

2

The Wright Flyer soars like an eagle compared to that thing.

3

Quadcopters are really maneuverable because they have a low moment of inertia, and can reorient themselves quickly when changing direction. This thing has a high moment of inertia and thus poor stability and maneuverability.

Maybe they could have individual rotors angled in different directions so that the whole thing doesn’t have to tilt to move in some direction. That would require a more sophisticated control system, though.

4

Another issue is that small rotors are really inefficient–but maybe it’s not a problem in practice:

  • If you get your 10 minutes or so and that’s all you need, who cares how inefficient it is?
  • Much of the small-rotor inefficiency is due to edge effects (high pressure air spilling past the tip). But with a densely packed hexagonal arrangement, some of these edge effects might actually go away. Not sure about this one.
5

Been done.