Tricopters are possible as tilt-rotors, where you have two motors spinning opposite directions on mirrored sides, and the third motor is on the midline in front or back and able to tilt left and right. This allows the vehicle to control yaw despite having an inherent yaw imbalance in hover. I suppose you could do any odd number of rotors this way, and more rotors would mean less inherent yaw imbalance if there's only one extra motor spinning a given direction.
Two is possible if the payload hangs below the props, and the props are able to independently tilt.
One requires collective pitch and still at least a small tail rotor to cancel out the yaw, as a helicopter, but that is far more complex and fragile. But it is the most efficient, as fewer larger props are always more efficient than more smaller ones.
So four just turns out to be minimum to have no moving parts besides the props themselves, and still have full control authority, and the control logic is more straightforward than tilt rotors and such.
I think this is incorrect, it's the somewhat unintuitive rocket pendulum fallacy.
There's a few demonstrated here (26-32 seconds in): https://www.youtube.com/watch?v=vBsyFj8bEJk
And here's one that uses the ailerons instead of being able to tilt the motors: https://www.youtube.com/watch?v=L2fgL97rgK0
I could've sworn Peter Sripol built one while on Flite Test too, but I can't find it atm.
Looks like the rocket pendulum fallacy is about expecting meaningful passive stability from the location of the center of thrust vs center of mass, but even 2-rotors need to be able to tilt independently (or deflect thrust) for active control.
Theoretically this could still work even if the center of mass was above center of thrust, but the tilting/vectoring responsiveness would need to be very high. These RC models at least move so slowly that the air resistance of swinging back and fourth really does help dampen oscillations passively, but they all still have active flight controllers that are trying to keep angular velocity at zero without control input.
> the air resistance of swinging back and fourth really does help dampen oscillations passively
But they don't swing back and forth like a pendulum (unless your PIDs are off) because of the center of mass, that is the entire issue with the fallacy.
Tandem rotor helicopters exist.
I was arguing against the COG of the payload, not of the existence of bicopters entirely.
As other replies note, 4 is the simplest arrangement mechanically and control-wise, as the control math is quite simple (just rotor speeds/torque) and the only moving parts are the fixed-pitch rotors.
The minimum, as seen in real (and model) helicopters, is either two (approximately) constant-speed rotors with swashplate control, or one (approximately) constant-speed rotor with swashplate control and one tail rotor, either with (approximately) constant speed and variable pitch, or with variable speed. At the scale of real helicopters, two rotors may often be more powerful and efficient (e.g. CH-47, V-22) but the size and weight of the gearbox needed to transmit so much power is a significant contribution to the weight and cost of the helicopter, and thus having a single main gearbox is much lighter and cheaper. The notable difficulties of shaft drive between multi-rotor helicopters, particularly with distributed engines (see a number of V-22 issues) strongly discourages helicopters with more than 2 rotors.
Of course there are 6, 8, and larger numbers of rotors used in actual drones. The advantage of more rotors is that redundancy to failure can be built in, and that rotor tip speed for a given lift can be somewhat reduced at the cost of efficiency.