El Reg: Pi-powered drones for disaster relief

[liz]

Just saw this, and thought you guys might like to discuss it!

http://www.theregister.co.uk/2012/06/12 ... _pi_drone/

That is a great project and I can see it will save some lives!

What I like for my personal Piiii in Space! project is that they're working on automated flight, something I'm probably not smart enough to work out on my own. Thanks for posting!

Just saw this, and thought you guys might like to discuss it!

Wheels (and a prop) but darn it still no sandwich.

Great idea, but will it get buried in red tape (looks like the plane's covered in it already)?
UAV regs are quite stringent, so will it ever get off the ground?
Seems like pi in the sky to me.


Hope they find a way round the bureacratic minefield though. Looks like a lot of fun to work on. The circumstances they'll be used in might allow for relaxation of the "usual" regs.

Haven't got time to dig into all the specs, but with a RaspberryPi, two Nanodes and an Arduino Mega, I guess the system should be fairly flexible, assuming they haven't made it too complicated.

I obviously wish them the best of luck.

This presentation about the plane/droneis really exciting. It looks like there are a number of great projects all wrapped up in this idea. They are also requesting technical help ... I think it would be brilliant if the community could help this project along either by signing up with Open Relief ... Although I cant see any obvious community around Open Reach except for this "outreach". I think it would be nice if the Automation, Sensing and Robotics section of this community could somehow extend itself around to help the Open Relief. I suppose that would involve someone from Open Relief starting the ball rolling on the forum with some specific questions or points but, to reiterate, I think it would be fantastic if that could happen.

As a pilot, just as I wouldn't be in any hurry to fly in any aircraft designed and built in a hurry, I also wouldn't want to be on the ground or in the air anywhere near something airborne built by people who aren't familiar with airspace regulations, not to mention the physics of flight. This is especially true in an environment where relief and search and rescue aircraft are operating - there is way too much that can go wrong just with their operations, and tossing this kind of wild card into the mix is highly inadvisable. There's a very good reason why we spend as much as we do on military, law enforcement, and search-and-rescue professionals and equipment.

The Arduplane software is, at best, only suitable for use in toys. There is nothing in it to ensure reliability of flight, and the entire flight system completely lacks any redundancy or graceful degradation when, not if, failures occur in hardware and/or software. Plus, there is nothing on-board to aid in collision avoidance, which is the number one risk in scenarios such as disaster relief. The sky is an even less forgiving mistress than the sea - and both routinely teach even the best professionals hard lessons every day of the year (see also Captain Sullenberger's comments on this subject).

Now, if they were to do simulations on the Pi using its wonderful GPU, I would fly and rest a whole lot more easily!