Single-wing Flight Based On Maple Seed Aerodynamics


The Samara Micro-Air-Vehicle is a product of over three years of work at the University of Maryland’s Aerospace Engineering Autonomous Vehicle Laboratory. The Samara is an applicant in the DARPA nano air vehicle program. Unlike the ornithopter we saw in July, this vehicle uses only one wing for flight. A small propeller on a rod mounted perpendicular to the wing provides rotation. The pitch of the wing is changed to climb, descend, or hover.

You can see a video of the flight tests after the break. The sound the Samara makes reminds us of classic alien invasion movies and the use of Verdi’s Requiem for the background music during flight tests (2:43) seems quite fitting. At about 5:45 there is some on board video footage that is just a blur of the room spinning by. This would be much more useful if a few frames per second were snapped at exactly the same point in the vehicles rotation.


[Thanks Waggy]

77 thoughts on “Single-wing Flight Based On Maple Seed Aerodynamics

  1. @rasz
    I was also struck with how awesome a linescan camera would be. Thanks for posting those camera specs.

    Regarding patents, this is one of those rare ideas that I would say deserves one. It’s oversimplifying to say that he’s “patenting nature”; how many natural maple seeds have you seen that can be remotely piloted and take video? 3 years of work went into perfecting the details of this design; it would be a shame if some chump could just rip off that effort.

  2. You cant patent a design (at least not in normal world, I bet USPTO doesnt care and just rubber stamps every garbage like they do with software patents), only implementation. There is nothing magical about implementing this thing in general. proves that having independent implementation (heh, they patented things like recovering angle of aircraft from the signal received by semi directional antenna instead of using gyro/compass, obvious stuff that should not get any patent on its own, but in US you just throw two things together and receive a patent).

  3. What is your definition of “design”? Are you saying that uses the same design? Then think of the ways that the implemention of this product is different. Seems that you could just as well say that these patents are on the unique aspects of implementation.

  4. Get the price way down and these would be
    awesome substitutes for clay pigeons.

    A high def panoramic camera could be made with a
    linear array of ccd’s with a slot aperture
    (reducing the need to transmit full frame info)

  5. A helicopter uses over a 1000 moving parts and has to be over a certain height and or speed to safely auto-rotate (dead mans curve). This thing has at min 1 moving part and 2 for controlled flight. As for the camera set the frames per second to match the rpm and increase or decrease the power to the motor to look left or right.This might seem like a pointless design now but so was the Write Bro’s flier (20ft altitude 120 ft distance and in a strait line hardly impressive)But look at airplanes now. Don’t kid yourselves THIS IS A MAJOR LEAP IN AVIATION

  6. @rasz

    Well, just the same construction as it is now, but without the servo controlling the wing pitch. So lateral control would work as it does now, but vertical control just by supplying more or less power. More power = up, less = down.

  7. A pod in the middle could be fairly easy to implement. It would need a gyro to keep the pod level, but I would imagine it wouldn’t take much to add a simple bearing of sorts around the pod. Controlling a camera within the pod would be the trick unless you were satisfied with always viewing from a fixed position.

  8. So here’s a thought for the vomitcam. The main problem, if you freeze the video on any frame, is not so much the high rotation speed, but the extreme motion blur. The apparently unblurred lamps in each frame are most probably caused by very short duty cycles, and each disk in the frame is the same lamp, smeared over the width of the frame.

    To solve this, a shorter exposure time would work, but at the cost of less light recieved, and still doesn’t really eliminate the motion blur; short exposure time != zero exposure time. The best way to deal with this motion blur is to eliminate the motion.

    Obviously the rotating platform could work, but that would probably eliminate one major advantage this design has, small part count and low weight.

    Since we’re thinking outside the box, suppose instead the camera were fixed to some sort of spring system, with maybe a piezo-electric or voice coil exciter to cause the sprung assembly to oscillate about the craft’s axis. there will be a brief moment when the camera’s angular velocity is canceled by the UAV’s rotation. An advantage of this is that the acceleration will be close to zero at the peaks of oscillation, so that there isn’t much extra rotation during capture.

    Such a system could be tuned to suit the camera, by adjusting the spring constant or mass of the system. No sliding parts are used, since the platform returns after each shot.

  9. The camera exposure issue for a cheap camera at high shutter is easily fixed by adding aligned images together. Line scan or regular, create a 360 degree image, some clever software stabilization control and a orientation sensitive HMD and you’d have an awesome little spy device!

  10. Dan K … up on your tech sights, saying that is standard for basically saying you must click the article to see the movies.

    Engadget uses that saying
    Slashdot uses it,
    Kotaku, etc etc

    Its not a joke you idiot <.<

  11. could also use a linear ccd extracted from a scanner- those are already fairly compact and with suitable software could generate a 360 degree image.

    interestingly with the modern accelerometers and single axis chip gyros this could be scaled down to pager motor size, using an electrostatic actuator for pitch control.

  12. @james – Thank you for that link! I’ve been trying to find that for months now! I remember seeing that thing years ago. The greatest thing about this design is that since it doesn’t rotate around a solid axis, it becomes translucent to a degree in flight. This makes it the perfect platform for a UAV reconnaissance device.

  13. @bothersaidpooh: The prototype demonstrated at 1:27 in the video uses a pager motor for thrust and + SMA for pitch. Of course SMA is a lot slower than the 10hz that would be needed for directonal pitch. I was planning on giving a voice coil a try if I ever build one.

    Also, the line ccd in a scanner sounds like a worthwhile avenue to pursue. Pricewatch reports new scanners to be had for as little as $25.

  14. umm…. nature created that design for seeds to fall down… not to hover upwards.. therfore i dont think this design would be better than a helicopter design, but could be usefull for other things..

  15. Instead of cameras, why not incorporate a radar system. The aircraft is already spinning, you could mount sensors to the edge of the wing that could create a 3D map of its surroundings.

  16. @AllYouPeopleWonderingAboutTheVomitCam

    Forgive me if this has already been suggested, ’cause I didn’t take the time to read every damn comment.

    I’m guessing that it’s fed into the software for the autonomous flight demonstrations.

    With the bright IR lights arranged at known places in the room, software could be written to extrapolate location and maintain altitude. Notice the green blobs traveling in parallel horizontal lines once it gets going, that’s an easy thing to monitor with some machine vision.

  17. Awesome concept, they did a great job designing this. But in my eye’s this mono blade less efficient than having two equally balanced blades. In order to balance out the mono blade, a large weight has to be used on the opposite side of the blade. This extra mass does nothing to help lift the craft, only to balance the wing. That same mass could be used to just add an extra blade on the opposite side and the extra mass would then be aiding in the extra lift of the craft. Much of the same reason why Cannondale’s Lefty fork never became mainstream in the biking community. It had to be over engineered to do the task of what normally was done by two forks. In this case, my opinion is that two is better than one.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.