A traditional quadcopter is designed to achieve 6 degrees of freedom — three translational and three rotational — and piloting these manually can prove to be a challenge for beginners. Hexacopters offer better stability and flight speed at a higher price but the flight controller gets a bit more complex.
Taking this to a whole new level, the teams at the Swiss Federal Institute of Technology (ETH Zürich) and Zurich University of the Arts (ZHDK) have come together to present a hexacopter with 6 individually tiltable axes. The 360-degree tilt in rotors allows for a whopping 12-degrees of freedom in flight and allows the UAV to fly in essentially any direction including parallel to walls.
In addition to the acrobatic capabilities of the design, the team has done some testing with autonomous control using external cameras. Their blog contains videos of their testing at various stages and it interesting to see the project evolve over a short span of nine months. Check out the video below of the prototype in action.
With Amazon delivering packages via drone and getting patents for parachute labels, UAV design is evolving faster now than ever. We can’t wait to see where this 12 DOF takes the state of the art.
Actually, a traditional Quadrokopter is NOT able to achieve all 6 DOFs. At least not indipendent. If a traditional Quadrokopter rolls to the right for example (a positive angle around the x-axis, which would be one of three possible rotations), he will also move to the right, which is one of the three translations.
So this new hexacopter is pretty impressive, as it is able to achieve those 6 DOFs independetly. But (!!) there are also Quadrocopters, with tilt-rotors – and they are also able to achieve the same thing, yet not so good looking :D
Think it might need some software upgrades, doesn’t look as smooth as the borg cube style one a few weeks back which had fixed motors strategically positioned.
Spot the raspberry Pi.
That thing is awesome. I wonder what kind of possibilities this opens. Might have really good use in cinematics, puts panning aerial shots to a whole new level. No need to worry about copter parts popping into frame, because the whole thing is capable of staying out of frame at all times.
They had a Pi *in addition to* the UpBoard?
(I did not see a Pi anywhere, I did see an UpBoard – the telltale large heatsink and backup battery stuck to the Ethernet port case are pretty obvious.)
Ah, then i was just mistaken by the eth, usb and HDMI arrangement.
just a few days ago a lot of folks were ragging on tri copters having the “added complexity” of tilting their tail rotors.
In a tri rotor the added complexity doesn’t add enough awesome to justify the complexity. In this case the awesomeness to complexity ratio is much higher.
Check out the E-VTOL market for further reading on passenger sized versions.
at 32s in the above video it looks like they are using a modified version of Gazebo and RotorS?
This is actually the GUI belonging to the VICON motion capture system.
But the team was supervised by people from the Autonomous Systems Lab, which is where RotorS has been developed ;-)
Really this short article is pretty uninformed, should hackaday articles be proofread?
Hexacopters aren’t inherently faster or more stable, there’s more to it. In the ideal world they should have about the same type of power-to-weight ratio as a quadcopter or most other multicopters assuming the same efficiency of each component and the same power / rotor area divided among 6 motors instead of 4, though in practice a 6 motor system will have a minimally, unnoticeably higher losses, frictions, etc. All in all there isn’t a whole lot of difference.
A hexacopter flight controller is not normally more complex. In the ArduPilot stack for example the C++ classes responsible for dealing with motor output for a quad and a hexa are subclasses of AP_MotorsMatrix which takes the number of motors and their coordinates and does the same calc independent of the number of motors, no added complexity of any kind.
12 degrees of freedom I believe is a marketing term, like the 10DoF china sensor boards? I’m not really sure but a robot in a 3D space can’t have more than 6 DoFs obviously. 12 actuators == redundancy. The cube thing from a few weeks ago used 8 I think and achieved the same number of DoFs, but it seemed better tuned in the video and by not using servos likely had more precision.
My understanding when it come to robots, is the each joint has its own reference frame, with its degrees of freedom relative to the previous joint. Which is where you get more than 6 degrees from. A flying vehicle on the other hand can only move with 6 degrees within a single frame of reference.
So basically, someone is going to build a Scorpion gunship from the movie Avatar before the end of the month.
Harrier-like? Or Osprey? One has propellers.
I’ve thought about having tiltable rotors on a traditional multicopter, but it adds a fair bit of weight and has very little practical use – it allows hovering at an angle, body leveling at speed, and that’s about it.
There is a “kind of” similar toy (http://s.click.aliexpress.com/e/M3jyVrn ) with variable angle, probably could be hacked for more interesting flying patterns
Just curious, why is something like this desirable? It looks cool, but is there a functional reason for the additional complexity and weight and reduction in prop area?
For reconnaissance, wouldn’t it be easier for the small camera to move? For payload delivery, this makes things harder. I guess it could now fit through narrower vertical gaps.
Basically this is an extenuation of my previous question regarding the benefits of a multicopter over a helicopter. The helicopter is more efficient (even with the tail rotor wasting energy) and quieter for a given payload.
“The more they overthink the plumbing….”