Hacker News Reader

Comment by chrisfarms | Hacker News Reader

Should this technique technically work for the failure of two motors too (assuming the remaining motors are opposite)? Or would it be impossible to build up the inertia for a stable spin? If motors could rotate, could it even fly on one? ... so many questions, maybe I should just build one.

View on HN

I'm no control theory expert, but I have a hard time imagining how you could control a quadcopter in 3+ degrees of freedom with only a single input (the speed of your single remaining motor) no matter how fancy your algorithms get.

However, as unwind mentions, the researchers in the article claim it's possible.

With one motor, if the vehicle is spinning, you could perhaps have a burst of increased thrust at a particular point in the spin and use centripetal force to move forward.

The article mentions that 'full control' becomes increasingly unlikely as you lose rotors, but partial control is still possible. You can influence how fast the quadrocoptor falls, reducing impact velocity or steering it to crash in a more favorable location.

The essence of the algorithm is that you spin in the horizontal plane. If you have enough power your single remaining motor can create a vertical airstream that will keep the altitude constant. The limit on spinning is the structural integrity of the system with all that angular momentum, the limit on flying is the frequency response of your control system.

A number of people have built and demonstrated so called 'half prop' helicopters based on the same principle. If you have ever seen an maple seed fall you can understand the principle, search for 'maple copter' on Youtube for some videos of them in action.

Example: http://www.livescience.com/12725-maple-copter-robotic-craft-...

That sounds like fun!

In the meantime, the actual linked-to article says "yes":

This new approach allows such a vehicle to remain in flight despite the loss of one, two, or even three propellers. Having lost one (or more) propellers, the vehicle enters a continuous rotation — we then control the direction of this axis of rotation, and the total thrust that the vehicle produces, allowing us to control the vehicle’s acceleration and thus position.

Reply on news.ycombinator.com