Autonomous Spaceplane Travels To 10 Km, Lands Safely 200 Km Away

Space balloons, where one sends instrument packages to the edge of space on a weather balloon, are a low-cost way to scratch the space itch. But once you’ve logged the pressure and temperature and tracked your balloon, what’s the next challenge? How about releasing an autonomous glider and having it return itself to Earth safely?

That’s what [IzzyBrand] and his cohorts did, and we have to say we’re mightily impressed. The glider itself looks like nothing to write home about: in true Flite Test fashion, it’s just a flying wing made with foam core and Coroplast reinforced with duct tape. A pair of servo-controlled elevons lies on the trailing edge of the wings, while inside the fuselage are a Raspberry Pi and a Pixhawk flight controller along with a GPS receiver. Cameras point fore and aft, a pair of 5200 mAh batteries provide the juice, and handwarmers stuffed into the avionics bay prevent freezing.

After a long series of test releases from a quadcopter, flight day finally came. Winds aloft prevented a full 30-kilometer release, so the glider was set free at 10 kilometers. The glider then proceeded to a pre-programmed landing zone over 80 kilometers from the release point. At one point the winds were literally pushing the glider backward, but the little plane prevailed and eventually spiraled down to a perfect landing.

We’ve been covering space balloons for a while, but take a moment to consider the accomplishment presented here. On a shoestring budget, a team of amateurs hit a target the size of two soccer fields with an autonomous aircraft from a range of almost 200 kilometers. That’s why we’re impressed, and we can’t wait to see what they can do after a release from the edge of space.

39 thoughts on “Autonomous Spaceplane Travels To 10 Km, Lands Safely 200 Km Away

      1. And winds aloft helped on the descent. This was why they chose the altitude they did. According to the stats at the end of the video, max (ground) speed was 212.5 mi/hr.

        1. From in the video, I thought that the initial “out of control” flight straight down just after the drop hit the top speed. That makes the ground speed much less, because most of it was “down”.

      2. So the landing site was 80km from the launch site, up 10km, sideways 80 km, down 10 km… where’s the missing 90 km?

        The balloon was going in the wrong way for a while, so that adds another 20-30km but still, it seems a stretch unless you count the bit where the glider went corkscrew over the landing site.

        1. Actually.. it’s worse than that.

          If you simplify the path, forget your corkscrew, forget wind. Just assume it went 10km up then took a straight line path angled down to the ground and landed 80km away. That path plus a straight line drawn from the launch to the landing sites would be a triangle. The path down is the hypotenuse of the triangle. The sides of the triangle are the 10km up and the 80 km back. The length of the hypotenuse is the square root of the sum of the squares of the two sides. So we get (80^2 + 10^2)^0.5 ~ 80.62. + the 10 km up makes 90.62. So we are missing 110km!

          No doubt the balloon did not travel straight up. No doubt there was twisting and turning in the path down. Still, to get 200km that makes the ‘extra’ distance traveled greater than the distance of the trip itself!

          1. It’s difficult for me to determine if you guys are trolling or not.

            Great-circle distance from A to B, at ground-level, is not 80 km, it’s approx. 143.3 km, depending on the method used.
            Predicted flightpath were 146.6 km and actual travelled distance ended up being 195.95 km (difference being small detour, corrections, and properly the way distance is measured/taking altitude changes into account).
            The plane was released from the balloon more than 80 km from B.

          2. Seriously? Why would you discount the GPS recorded flight path and calculate distance traveled based on straight lines? Where else in the world is it done that way?

  1. The plan and balloon didn’t travel in straight lines. The specifics are at the end of the video.

    That is awesome and beats the hell out of you payload landing jnsome random location though in Australia my understanding is it is not legal to drop a plane from a balloon. Other wise I would be following in their footsteps. No I would hav retried it already…..

    Congratulations on their fine effort

  2. I like the autonomous glider bit, and taking the trouble to plan the flight so they wouldn’t just lose it over the ocean, but calling 10,000 m “the edge of space” is a bit of a reach. You realize that’s where airliners fly, right? Which reminds me…

    Also, it’s probably a good thing it didn’t go much higher than that: hand warmers work by oxidizing iron powder by exposing it to AIR, so once you get up to some serious altitude, they kind of peter out.

    1. The single use ones might use iron powder; the reusable ones use supersaturated sodium acetate solution crystallising. I wonder if it’s possible to have a servo click the little metal disc on one of those so it only starts releasing heat on a time delay…

      1. Good point – I wasn’t thinking about those. I’m sure you could give a servo enough leverage that it could operate the clicker. I don’t think the sodium acetate ones have as high energy density as the iron ones, though – burning things generally produces more energy than fusing things. ..looking it up now: sodium acetate latent heat of fusion = 252 kJ /kg, iron oxide (III) heat of formation 824.2 kJ/mol / .16 kg/mol = 5150 kJ/kg. I’d probably just put everything in a tape-sealed block of styrofoam so the heat dissipated by the computer and flight controller would take care of that. I’ve seen other projects that did this successfully.

        1. It’s a lot easier to heat electrically, if you need better thermal control than you get just from insulation and the natural power dissipation of the electronics. The favorite batteries for this sort of thing ( have an energy density better than 1 kJ/g, and it’s quite easy to make a thermostatically controlled heater.

    2. The cold isn’t as much of an issue as one would think. As the air thins it becomes less conductive of heat. The electronics are pretty good at self heating and its only spending an hour or so at high altitudes. Insulation is all that is required. Over heating might be my bigger concern.

      On the balloon launch our group did last year the electronic were all in a box made of 1″ EPS foam, we had no extra heating and the box never dropped below freezing. We had a GoPro, RPi /w camera, spot tracker and a 300mW APRS transmitter onboard. We powered it from a lithium ion USB pack. We reached over 70,000ft.

      Unfortunately operating an unmanned aircraft beyond visual line of site is illegal in most countries. This breaks a number of other laws as well.

  3. If it had a wider wingspan, it could probably fly even further, but perhaps long, fragile wings wouldn’t fare good in the strong wind during the ascend. From the reconstructed path, it was not aiming for the record trip length, instead it was supposed to land on a pre-determined spot, to rendezvous its awaiting land crew.

    This is too powerful feat – doing it is probably going to become illegal, or highly regulated.

    1. It’s probably already illegal. No mention in the description of what precautions they took or clearances they got for operating this in class A airspace (everything above 18,000 ft, because feet is how we measure altitude around here; about 5500 m for everybody else).

    2. Regarding wind, the plane was tethered to a balloon that was being blown by the same wind, so the airspeed of both would be close to zero on the ascent – the only thing the wings would have to endure is any turbulence along the way. Same thing on the descent – according to the video, the maximum speed was 212 mi/hr. Clearly that was ground speed; that plane probably couldn’t tolerate 60 mi/hr of airspeed.

      But as you say, they were more interested in autonomous flight to a preset destination, and they chose an altitude where the winds aloft did most of the work of getting it there. All they would have gained with wider wingspan would be a lower descent rate, which would just burn more time and battery power on the descent for no good purpose.

      1. The other thing to remember is how thin the air is above 10km. They would have little to no control at higher altitudes. At higher altitudes the parachutes on a regular payload do little to nothing. It tumbles and drops like a rock until it gets below something like 10km. So you can’t make up for drift with a glider until it falls back into thicker atmosphere.

        On my last balloon launch we didn’t have the GPS in the right mode so we don’t have any data above something like 30,000ft. We got APRS back when it dropped below that point and it was dropping really fast and then slowed down. We did get lots of pictures of it tumbling around at altitude and a log of pressure(altitude).

    1. Wow… achieving basically the same stunt with a 25MHz DOS computer and a 35mm film camera? Give those guys a Raspi and a some modern GPS and telemetry modules, they’ll probably go to the moon :)

    1. You mean something like a Rogallo wing?
      Seems like it would still be considered a glider, just with a non-rigid wing. I suspect that it would fall under the same laws as one with a rigid wing, but I could be mistaken.

      It sounds like fun, so it must be illegal.

  4. So what part of this happened in space? Seems to me that buoyant flight has kind of a fundamental problem reaching vacuum, and maybe you should quit bullshitting about balloons going to space.

    Unless it’s a balloon on Mars or Titan or something, which got there by riding a rocket to space — THAT would be a space balloon.

    1. The article uses ‘edge of space’.
      But correct, space starts at 100km up so it’s not a ‘spaceplane’ if it goes up 10km, or even if it had been the initially planned 30km.

      Also, normally the speed of a plane is measured in ‘knots’. (although I prefer km/h) so it’s a bit amateurish perhaps to list km/mi instead? Not that it matters much.

  5. Congratulations!
    How about law? Glider like this is qualifying as a drone or falling stratosphere probe? In 2 months I’m sending mine glider and have no idea how to make it legal :)
    Greethings from Poland! :)

    1. Balloon – you submit a NOTAM at your local air traffic control agency and have it approved…that’s it. If they’re not a bunch of cunts, they should even be able to help you with that.

      Autonomous glider in controlled airspace – you can’t. Until the laws change, it will be illegal in all of EU airspace :(
      So, either don’t tell about it too much, or leave the glider below what’s considered controlled airspace.

      1. OR – find some contacts in the military and try to persuade them to allow you to use the anti-aircraft firing range you have in Poland on the Baltic Sea beach :D
        (I know there is one because the CZ army has to train it’s AA corps there, our tiny republic doesn’t have a large enough military polygon to do it safely here)

        1. I think AKA the A think about military area near Ustka (i live there).
          Konrad, can you tell more about your project, it sounds very interesting and i think i could help you,

  6. We did an autonomous glider via balloon a few years ago. Somehow the conversation got on the balloon bombs the Japanese used in WW2 and things naturally went to how it applied to a glider platform. It would be a very cheap cruise missile of sorts. The FBI was knocking on our doors a few days later and requested we remove the materials and conversations from the web. With the glider autopilots available now it would be easier and even more accurate.

    1. I was looking at doing something similar myself a few years back and I came across an FAA ruling that it is illegal to launch a aircraft (of any size) from a balloon. So I nixed that project. I’m trying to find reference to that now but not having a luck.

      1. There are lots of things the regulation say you can’t do, unless you apply for and get a waiver. We had a NOTAM issued by the FAA for the airspace when it was flown. The rule is there to prevent things from being dropped into class A airspace that could present a hazard to flight. You can get waivers. It’s not easy, and it requires everything to be thoroughly documented for FAA review along with the waiver request. Don’t expect it to be quick. It can take well over a year.

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.