Came across this in the Science Museum London as a development by Manchester University. It looked like the motors are fixed at this angle - Why? Anyone suggest any benefits for this arrangement - faster yaw control? greater stability? more controllable?


E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones


  • There is a picture of the caged version in this http://www.science.mod.uk/codex/documents/Codex_issue2_GC_Supplemen...
    from 2008 which seems to have three props angled down and three up.

  • That's the one - Bill was my lecturer in Aerospace Systems. Great guy. I've seen a version of this displayed in our lobby with a roll cage off sorts, for that rolling motion. The bonus of being able to apply a thrust in any direction... I think they abandoned the gripping- feet idea in the end, or at least decided not to actually try building it. That post must be a little old, BTW - I first saw Tumbleweed in a lecture about two years ago.
  • Thanks Nicholas, knowing the name Tumbleweed brings quite a few hits including this:
    "The Tumbleweed was designed by a team from Manchester University, led by Dr Bill Crowther, senior lecturer at the School of Mechanical, Aerospace and Civil Engineering. He remarks:
    'The idea is pretty new. This design only dates back around 10 months, and we only got a patent granted last week.
    'The problem with existing Unmanned Aerial Vehicles is that they're fairly unmanoeuvreable, particularly close to the ground among buildings in gusty winds.
    'Instead of having to turn and tilt in order to move - like a helicopter - the Tumbleweed remains level and relies on very rapid variations in power to each of the six propellers - so it can generate forces to move in any direction.
    'It will fly but it will also roll along the ground which is a far more efficient way of moving, so we have longer endurance."

  • Former Manchester Uni student here - the UAV is called Tumbleweed, and was part of a project being run by a lecturer of mine. The purpose of the weird rotor angles is quite cool, actually - by combining different speeds of the various rotors, it can apply a thrust in pretty much any direction, which means it can hover any way up. The legs are on the bottom and the camera is on the top, but it can quite happily flip over to get a better look, and if you were to give it some kind of grippy landing gear, it could latch onto lampposts or the underside of bridges to save battery.

    It was entered into an unmanned vehicles competition run by the UK Ministry of Defense, actually - some clever googling would probably turn up a little info, and maybe some clips of it in flight. I've seen it in action, and it's fairly cool.
    This domain may be for sale!
  • I have toyed with angling the props at 45, 60 and 90 degrees and they all worked very well. I believe the distances and size of the props plays a very important role.

    V-Prop60 - Park Flight Test
  • From what I understand, the purpose this class of thrust orientation is improved attitude maneuverability.

    I've seen similar designs - all essentially involving (seemingly) randomly oriented rotors/thrust mechanisms extending from an axis point.

    Applications would involve sensors and payloads requiring 360 in xy&z; complex fluid or space borne operations, etc

    In a quad, yaw motion is realized in the direction of the induced reactive torque; sideways motion - by rolling and increasing/decreasing the left/right rotor thrust. I think this hex does the opposite.
  • Sadly the exhibit itself had been removed to be used at a meeting. The point of the display and audio visual was to invite comments about the use of such devices for police surveillance http://antenna.sciencemuseum.org.uk/should-the-police-use-flying-cctv/

    It has resulted in 905 comments.

  • Probably why its in a museum and not out flying.
  • Can't possibly be a good thing; any inefficiency simply reduces the payload/flight time.
    so let's imagine that the confluence of the two streams is reasonably efficiently directed downward, certainly changing the relative rpm would change the effective thrust angle - so it might just be possible that this configuration is capable of navigating in 3d space WITHOUT changing its attitude.
    It seems the neutral torque pairing algorithm for a steady attitude hex is interesting:

    I notice that each arm contains two props with are aligned - let's assume that they are contra-rotating.
    so if we assume stable flight, then lateral movement can be achieved by increasing thrust accross a pair of contra props in the desired axis.

    We might achieve level flight by using torque differentials across each contra-pair according to its axis. That leaves yaw stability:

    For this perhaps we group all left-oriented props into one group and all right-leaning props into another basket, and apply a Yaw feedback to each group accordingly - this would undoubtedly lead to leveling problems - but it might resolve to a solution.
    See related links to what you are looking for.
This reply was deleted.