Nothing that I have mentioned can be regarded as NASA level technology. Most of it has already been done in the RC community, only on separate projects, and never combined in a single effort. If you google around, you will find a good number of RC builders who have managed to build UAVs that can fly for 2000-3000km, without using any hi-tech methods. Sure they did some mods to the engines and communicaions, but those were nothing that the experienced RC builder couldn’t do. I have also found a couple of people who were talking about building wet-wing designs, tough never found any reults.

Its getting a lot easier, and cheaper, to use compisite materials such as carbon fiber and epoxy to build RC planes. Get a descent CNC router (you can build one for under $1000 if you have some tools and good knowledge about CNCs, again google around to see how many DIY CNC projects have been done) and you could pretty much make any shape or size art you want, with virtually any material you can get access to.

As for control software, there are many RC autopilot projects, some are open source, others are commercial, but still, available to the average RC builder. Check out sourceforge to see how many people are working on that. Also take a look at the Paparazzi autopilot project which has some very nifty features. Look at gumstix who sell some powerful and increadibly small xscale based computers running Linux. A 400MHz Xscale can do give a reasonable amount of processing power, considering that you don’t have a GUI overwheliming it, and that its running Linux, which runs a lot faster and provides a lot more flexibility than WinCE does. The rest is all software that can be done with a couple of descent programmers, not rocket scientists. There is very little, to no electronics design invilved.

All of the Satellite phones I have worked with in the past provide a serial interface, which operates using standard AT commands, the same goes for a good deal of mobile phones. If you dont like to use phones at all, google around to see how many types and models of wireless modems are available. Some use AT commands, others provide a seamless wireless serial interface that is redicolously easy to interface and use. As far as electronics are concerned, there is a goodchance that the biggest prblem you will face is finding or making an efficient antenna for your wireless modem that is small and light enough to fit on board the UAV.

As to your ideas, well those aren’t the things an average person with some basic electronic knowledge can build. You might as well try to be NASA. I can build a car sized blimp, heck they probably sell it on Ebay. A ballon requires no power to stay up in the air, so night is not really a problem. I think the old derigibles were able to make trans atlantic flights with an 8 gallons of gas and a small 50hp engine.

In contrast, a RC plane will stay in the air for 10-20 minutes with a gas or glow engine.

The glider attached to a weather ballon was cool. I think the germans attached a biplane to their derigibles. This is probably something you can build. Although it sounds like your interest lies with planes rather than ballons.

Well it was nice banter,
Jonglee1977

I did some reading about that, mostly studying the feasability of a high altitude (not so) mini blimp that carries some 5-6lb of payload plus all the power, control electrronics, and communication equipment. I was more thinking of integrating a satellite phone to relay telemetry and recieve commands through SMS, which would reduce operational costs considerably. However, there are a number of hurdles to overcome. First, you need to carry enough batteries to power the blimp during the night. Second, the extreme low temperatures encountered at such altitudes. Third, the relatively uncontrollable, and unpredictable weather conditions at high altitudes. At over 60,000ft, you will encounter quite high wind speeds, which could stress the blimp. Finally, while it has already been done, though not for long periods of weeks, the use of such a vehicles are rather limited. You might want to google jp aerospace, or take a look at this guy, who is working on a mk2 of that project.

On the other hand, a long endurance UAV does have a wide variety of uses, from the military, to survelliance, and even as a relatively safe and low cost test platform for new technoligies (no humans on board, so there are no lifes at steak of something goes wrong). One of the problems with long endurance UAVs is supplying enough electrical power for the electronics. Sure its possible to use an alternator connected to the main engine, but that adds a considerable amount of weight, and draws a considerable amount of power from the relatively small engine that runs the UAV. By using solar panels, the need for an alternator would be eliminated, at least during day missions. Another use for the solar panels could be to supply power to run an auxilliary electric engine, which would enable the UAV to travel faster, or reduce the load on the main engine, by reducing or even eliminating the need for the main engine to produce thrust and just run at idle to produce power for the flight electronics.

Those were just some of my thoughts on how to approach the problem of designing a long endurance UAV. As I said in the original post, I have another bunch of ideas regarding ground control, navigation, (real) automatic landings and takoffs (rather than some form of controlled crash landing), laser guidance (for droppable payloads), among other things.

One can even devise a miniture helium tank to replensh the slow leak of the blimp.

Of course you’ll pay a lot of money for roaming charges.

jonglee1977