An update on my project (please see my previous two posts).
See http://www.youtube.com/watch?v=lhdR2Iesfuw
This is a screencast of APM2 running in HIL mode with the X-plane RC plane in an 8kt wind over a simple course, takeoff, circuit including loiter, land, fully auto (radio not touched). In JSBSim/SITL results are actually better than X-plane HIL which has poor resolution and slow sensor update.
If anyone wants to try this firmware, in X-plane or in HIL or SITL *please* let me know. I would like testers.
I can build it for APM1, ardupilot-mega and APM2 boards or of course there is the desktop version for SITL
I'll upload the source to a git clone soon as well.
My plan is to fly a skyfun and/or skywalker with it when the weather relents.
features so far:
Same base code as ArduPlane, but new navigation, stabilize and autopilot PDEs
FBW-A, B
STABILIZE
GUIDED
LOITER
RTL
WAYPOINTS and the normal mission planner stuff
geofence
mode switching
Auto-trim - boot with elevator stick at top of the range and plane will fly a short routine to self-tune as soon as you hit AUTO mode.
Presets - presets for several different kinds of UAVs
More free memory
Trig lookup tables to free up cpu
50hz navigation
GPS extrapolation
Comment by Justin on January 25, 2012 at 6:40am The sharp eyed may notice there was no track following between waypoints 6 and 7
this was not because of wind or anything, but because there was a speed change command between waypoint 6 and 7 this the navigator did not have a "from" bearing on its way to 7, so it just made a bee-line.
I've fixed this, but could not be bothered to re-do the screen cast.
There is a bunch of stuff I can say about LAND, TAKEOFF and Auto-trim, if anyone is interested just ask.

Looks realy nice, I would be interested in testing the SITL setup, aswell as one for APMv1
Comment by Jeb Bailey on January 25, 2012 at 12:17pm Please, do post about the auto takeoff/land.
Comment by leonardo.bueno on January 25, 2012 at 12:34pm Justin, I'm interested in testing it in SITL mode too.

Hi Justin, I'm back from my conference now and I'd love to have a look at your work. Could you point me at your patches?
Comment by Justin on January 26, 2012 at 12:42am For SITL tests right now all I have is an OSX .elf, so unless you are on Mac I can't post that.
I'll post an APM1 and APM2 .hex file shortly, then upload the code so anyone can pull it down and build it.

Comment by Michael Pursifull on January 26, 2012 at 12:03pm Justin - would't this look even nicer with a Bixler/EasyStar X-Plane model?
http://diydrones.com/forum/topics/x-plane-arduplane-bixler-easystar...

Comment by Michael Pursifull on January 26, 2012 at 12:12pm One item that has concerned me, between my HiL and real-world flights, is that where I fly, the winds are not 8.75 at 159.8 degrees, but rather they vary from 0-8 knots, average around 4-5, and change direction slightly from second to second (though generally from the same 5-10 degree source.) Maybe I need to fly higher? I am curious about your thoughts on this.
Hi , what is SITL?
and in what this project is different then the one that shows you how to connect the APM to the X-Plane

Comment by Michael Pursifull on January 26, 2012 at 1:35pm Software-In-The-Loop
His project, unless I misunderstand, is a modified version of the firmware to handle tracking differently, as well as several other modifications/features, which he lists in the original post.
The HiL (hardware in the loop) is used to make development of these features faster and less expensive (when you crash your plane, you don't have to rebuild it, you just reload it...and you can create predicable flight conditions on demand, and not wait three days for the rain and wind to pass)
The SITL is mentioned because of the difference in experience he's had between HiL and SitL. When you are simulating the sensors, pushing the data across your X-Plane, your operating system, in and out of the APM.... these things take time. Time is sensor lag, basically, and can dramatically change the flight experience. That being said, it remains an invaluable tool for development and for training.
Season Two of the Trust Time Trial (T3) Contest has now begun. The fourth round is an accuracy round for multicopters, which requires contestants to fly a cube. The deadline is April 14th.51 members
682 members
184 members
8 members
130 members
© 2013 Created by Chris Anderson.
Powered by

You need to be a member of DIY Drones to add comments!
Join DIY Drones