limitations in autopilot and GPS soft/fimware

[Sampey]

Hey all, havent posted in awhile, ive come across a technical issue in me slow but sure progress towards combining FPV and amateur space photography. Actually ive come across many, but most are resolved with simple solutions (greater Mah batteries and bigger SD cards in cameras for 3-4hr video recording, Long range UHF control and camera links for flying something 34km or so up).

Heres my pickle: all the OSD/Autopilot systems ive looked at have altitude and range limits (typically 10km radius from launch and 18,000ft) im going to assume exceeding these results in return to base failsafe triggering or total fail.
Seems to me that while FPV is possible for high altitude flight, its kinda pointless, what with control lag, and the long boring flight up attached to a balloon etc. I figured, what if we replace the usual amateur space mission, camera-in-foam-box-with-gps-beacon-and-parachute system for return from balloon launch, with a small autonomous glider instead.
The idea would be it flies home or near home allowing a manual landing instead of just dropping wherever, this makes it safer than the parachute method not to mention way cooler and easier than travelling long distance to recover your camera and the all important images and video.
Simple in theory, in fact a completely autonomous system could be much smaller and lighter than your average FPV plane, perhaps based around a lifting body design.
The only real issue so far with the design concept is the limitations mentioned earlier in the currently available autopilots and such.

The task is simple really, once released, the craft would need 3 axis levelling via gyros or accelerometers, and GPS waypoint following or position hold (as in circle continuously until you land, or manual control is established) this stuff is standard for even cheapo autopilot gear, but again the range and altitude thing, whats with that?

Assuming this could be made to work, in later develoment id like to have hybrid rocket power, for a short burn at altitude waaaay above jets and so on, that due to the thin air might well allow the craft to break the sound barrier on the way home. Some of the larger model rockets can do this in a vertical climb from the ground so im thinking itd be easier at 80 or more thousand feet up.

Tell me it wouldnt be cool to have a micro lifting body hanging in your workshop that you could say had flown at 1000kms an hour and been to the edge of space?

So! Can we reflash the gear? Build an unlimited arduino type system from the ground up? Ideas anyone?

[Sampey]

some views but no answers, well, fortunately i found my own, ardupilot mega 2.5 has no such limits and im told has been used upwards of 100 000ft already, mint
so thats the last on paper tech problem solved, now to continue physical model development and test flying, hopefully i can skite about pulling this off sometime early next year.

[mkuvy]

And if you are on a budget, Crius AIO Pro will do the job as well.