Our High Altitude Balloon (HAB) project aims to return the electronics on board a weather balloon autonomously instead of drifting away on a parachute. My Project Log series aims to provide short sweet and to the point updates to said project. Heres the most recent one:

Heres some background:

Our High Altitude Balloon (HAB) project aims to return the electronics on board a HAB autonomously. Traditionally, weather and HAB's return electronics safely to the ground via parachute, but this obviously steers waaaayy off from the launch site and takes hours to find. We aim to cut down the risks involved by just bringing everything back via flying wing.

The flight plan is as follows:

1.) Climb to 100,000 feet under balloon
2.) Have balloon explode at highest altitude and have small parachute bring down the plane from 100,000 feet to 30,000 feet. Lack of air pressure makes the void from 100,000 feet to 30,000 feet an uncontrollable abyss for the glider. Small parachute is used to decrease terminal velocity.
3.) Airplane releases itself from flight line and glides towards destination.

My Project Log series aims to provide short sweet and to the point updates to said project. Heres the most recent one:

This week I try to start making progress on the flight controller. Starting off with the original program code written by Kemal, I try to fix things here and there to smooth out the glide.

Next video will explain the program a little bit more, and will be of trying to achieve a successful autonomous straight line flight, and trying to incorporate turns.

Side note, you guys know of any other arduino flight controller projects i could pick up on and learn from? Computer science isnt my strong suit.

Previous uploads:

Project Log 1: Introductions

Project Log 2: Base Flight Test

Project Log 3: Q&A about design choices

I'm trying to do this whole project from the bottom up DIY, so i'll be keeping an updated, weekly project log series going! Let me know what you guys think of the project so far, and drop any feedback if you have them!

E-mail me when people leave their comments –

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

Join diydrones


  • 4 years ago something similar


    FPV to Space and back
  • Awesome! Thanks for the links Andy - bedtime reading for at least one evening.

  • Ok here is one of the papers we released:


    and here is the dissertation on the ETH Zurich webpage:


    AMT - Return glider radiosonde for in situ upper-air research measurements
  • Any links to your PhD dissertation Andy?

  • @Andrew Rabbitt

    No problems at flying at 100'000 feet have done it during my PhD project building a meteorological drone flying back expensive Instruments of up to 2kg weight

    The only challenges are to conquer the cold and the high flying speed once you release the plane form the balloon.


  • More or less, Tarik. 

    Delta wings have the advantage of vortex lift (Google it) so they potentially suffer less from low Re effects which could help expanding their useful AoA envelope, plus they are nice and stiff for a given wing area.

    In terms of control area size, whatever gives you enough control authority at sea level will work at altitude since your IAS is going to be the same in both cases.  The only thing altitude will do is reduce your AoA (and therefore control deflection) window of operation.

    Interestingly, your controller gains will probably be an inverse function of TAS rather than IAS so will need to be modulated accordingly.  I think this is an inertial effect

    I did a bunch of simulation work on a similar project using X-plane - you can read about it here

  • I was under the impression that roll and pitch would operate like torque. Put the control surfaces further out from the pivot point and you have more authority. Increase the surface area, and you would increase your force or pressure acting at that point, thus increasing your torque. The problem is, just with control surface area this small we're barely making it at sea level. I'm honestly concerned about 30,000 feet but with speed that will smoothen out. So I guess i'm not entirely understanding what you mean by "control surfaces dont need to be any larger". 

    And yeah we got that piece of advice earlier on too, there was even a space shuttle like approach to this whole thing. Just uses the entire 100,000 feet as fuel to get back to home base. It feels like a better concept honestly.. I think i'll be making multiple test bed aircrafts to test, and once we have the flight controller ready i'll test them all at the 400-1000 feet testing stage. 

    Is the space shuttle approach similart to what you meant by the small delta wing plan?

  • IAS will be about the same as at sea level, but TAS will be large and you'll need to check Mach numbers too! :D

    Control surfaces don't need to be any larger but you are operating at very low Reynolds numbers so aerofoils are prone to high drag outside a small-ish C_L window due to separation bubbles forming.

    A small delta wing-style aircraft might be the best but it looks like you're going to be ambitious with your L/D - 15 will be a challenge to acheive,  Trade some of this off by covering distance from a higher altitude.

    Also, high speed is good to combat wind conditions, so higher wing loadings are possibly in your favour.

  • You definitely could! The Sr-71's flown pretty high, and a bunch of airplanes have flown above 100,000 feet. Just increase control surface area + airspeed, and at some point you'd definitely be able to control the aircraft.  It isnt if we can or if we cant, it's more of "do we really need to?".  Since our glide slope is high enough to cover 90 miles at 30,000 feet in a perfect world, we probably don't need to. Plus, there's a difference between an aircraft surviving a certain climate vs being able to fly in it. We would need to increase control surface area, which adds unnecessary drag for when the aircraft does not require that kind of roll rate and pitch authority at lower altitudes. Also, stiffening the aircraft to fly at a certain speed would also add weight. Just a couple things to think about. I don't think there's been a controlled drone flight at 100,000 feet though.. if so, it would be an interesting record to break once we have the system ready as a separate project. 

  • I disagree that you can't fly at 100k feet.  IMO you can if you get your aerodynamics right but the flight envelope shrinks a lot

This reply was deleted.