We're obliged to Linus Penzlien (seen above in boffin pipe mode) and Andrew Tridgell for all their hard work last week working on the custom APM:Plane parameters/commands for our Vulture 2 spaceplane.
We were finally able to get the aircraft strapped to the van roof and out for a systems test, and to run the MAV_CMD_NAV_ASCEND_WAIT command, while makes control surfaces "wiggle" every X seconds. This is to prevent the servos freezing up as the Vulture 2 ascends to launch altitude.
The Pixhawk worked fine, and here's a fetching graph of GPS and airspeed sensor speed during our 20 minute test run:
We do, however, have battery issues, and we're certainly going to need more juice. Full details on the test run can be found right here.
Comments
It is for sure one of the most advanced and exciting projects and group I've ever had the chance to worke with.
Big thanks go to theregister.co.uk, 3DR, Tridge, Lester and all persons that where involved over the years!
I am looking forward to continue working on the project and of course the fun things for what the project is famously known!
Don't fret, you'll get plenty of warning.
I think I will keep half an eye on http://spacenear.us/tracker/ as well
The launch date is imminent. We'll have some big news regarding that later in the week. Watch this space...
Ah good show I think pitch is going to have to be tracked at a heck of a rate against best speed to fly. So much learning going on in the falling process. Refresh my mind, when is the target launch date again?
Gary - we don't test gliders! You'll have to wait and see at just what heck of a speed it arrives. I believe there is some serious autotune skulduggery afoot...
Has there been a balsa or foam test glider?? Is the landing area huge and very soft??? I have a hunch it will arrive at a heck of a speed. Oh another question will it be running auto trim routines as it descends? Its all just too exciting get flying already.
@Philippe: Presuure sensor don't work well at these altitudes. We've got the Pixhawk sensor on board but will be mixing towards 100 per GPS as the plane goes up, and then reversing that process as it comes down.
As far as I'm aware (and I don't have the exact details), the aircraft will arm the autopilot some 20-30 seconds after ignition detect, or when it detects its already falling (whichever comes first). Once control has been established, it'll then perform a descending spiral towards the landing point, adjusted for a straight approach in the last 1,000m or so. Simple as that ;-)
We're still working on the parameters, according to the flight prediction data from the design guys. This includes various glide slope calculations at different altitudes.
@Linus,
Looking good:-)
Regards,
TCIII Autonomous Vehicle Developer
Hey guys,
first things first: I absolutely love your project it seems to be really fun and also very well made!
However I have a couple of questions: Which differential pressure sensor are you using?
Also how is the mission planned to be? From what I gathered, you will have a rocket-powered ascend phase after a balloon ascend, then you hook off your aircraft and sail down to a gps position? Will this be more like a gps boomerang in which you glide to the nearest point? When will you try to stop your free fall, thus when will you attempt to pull out of the straight dive?
Sorry for all these questions, but I am pretty interested in your project :)
cheers,
Phil