# How can this single-prop drone maintain stable (rotating) attitude and position?

The short BBC news item and video The drone designed to fly on one propeller shows the flight of something that looks like quadcopter missing three of its propellors and motors. It continuously rotates around a vertical axis through its center of mass tilted at perhaps 30 degrees, and the single propellor can provide different lateral thrust in different directions by (I assume) quickly modulating its power throughout the rotation cycle.

This seems to be only a demonstration of an interesting effect at this point in time, but I'd like to understand how it can retain both stable position and attitude with only one degree of freedom (speed). Yes the attitude is rotating, but that motion appears constant.

So far I haven't found anything to read about how this can actually work. In the video there is a suggestion that this has been calculated and equations are briefly flashed on the screen.

The work is done at ETH Zurich and the Flying Machines Arena.

How can this single-prop drone maintain stable (rotating) attitude and position?

I'm also curious if this is a 'new' effect, or is this a version of something that has been previously demonstrated or at least discussed.

above: GIF made from screenshots of BBC video found here.

above: Screenshots of BBC video found here.

http://flyingmachinearena.org/wp-content/publications/2016/zhaIEEE16.pdf

First, the parameter space (so, location of propeller, mass, etc) was searched for a relatively stable solution.

Then, stability is achieve through active control, so indeed by varying propeller thrust rapidly. This is done in a few cascaded control loops; so an 'outer loop' that has a position setpoint, which is then translated to a desired attitude. This attitude is then achieved by varying thrust throughout the rotation.

• This is exactly what I'm looking for - thank you! "Abstract— This paper presents the 'monospinner': a mechanically simple flying vehicle with only one moving part. The vehicle is shown to be controllable in three translational degrees of freedom and two rotational degrees of freedom." In addition to the control system, I'd like to figure out from a dynamics point of view how it actually controls all of these degrees of freedom.
– uhoh
Nov 9, 2016 at 9:12
• @uhoh All of that is in the paper. Of course, a background in control engineering is preferred to understand the specifics. In short, they use a linearized system matrix to get a LQR feedback. I could explain everything in this answer, but that would probably come down to copy-pasting their paper which I do not want to do. Nov 9, 2016 at 9:19
• Ya I understand your point. I can see that vertical translation comes from average speed, and lateral translation comes from modulation of the speed with period $1\omega$ where $\omega$ is the period of rotation of vehicle around the vertical, so really I'm just wondering if the "control" of the other two rotational DOF claimed are simply passive, gyroscopic stability inherent of a rotating mass, or if there is something active there. In what way is it actually "controllable" in two rotational degrees of freedom. I'll read the paper and if I can't figure it out I'll ask separately. Thanks!!
– uhoh
Nov 9, 2016 at 9:56
• I think the spinner is stabilized mostly due to the drag of the trailing 'branch' of the Y shape, which prevents the vehicle from 'flipping' along the axis that is attached to the propeller (say, the bottom of the Y-shape) Nov 9, 2016 at 10:05