Autonomous Senior Telemaster
I have just posted a youtube video showing my autopilot flying a senior telemaster. I included the take off and landing (both manually flown) because I thought it looked cool.
Here is the youtube link (I don't know how to embed the video here in the forum area.):
Here are a few details. Aircraft is a Senior Telemaster, electric powered. An 8000 mah (5 cell) battery yields 30+ minutes of flight if I slow cruise at 30 kts. The avionics are powered by a gumstix verdex running the main loop at 50hz. The IMU is a vector nav. We are running a 15 state kalman filter that produces a good estimate of roll, pitch, and yaw by combining the inertials with the gps data.
Recently I have been working on code to estimate the wind vector and true airspeed. I can use these estimates to plot stable up wind courses and roll out of my turns in time so that my ground track lines up with the target ground track (even in strong head winds or tail winds.)
As you can see from the windsock on the landing, the winds during this flight were gusting up a bit. On the next flight I was recording winds in the 14-16kt range at altitude. The aircraft was bouncing around more than the clip I show here, and at times I was crabbing more than 45 degrees to maintain my target ground track.
Replies to This Discussion
-
Permalink Reply by Curt Olson on -
One other comment. I have taken the autopilot module out of FlightGear (which is pretty sophisticated and stable and very robust to things like windup) and used that for my actual UAV autopilot. One cool thing about this chunk of code is that the entire cascading series of PID loops are defined in an XML file. You can specify the input parameter, the reference parameter, and the output parameter in a very generic way, using any internal value of the autopilot ... so the result is a very very flexible system.
I've shown some of my flight video to other FlightGear developers and they comment that they can see the UAV flies very much like FlightGear airplanes under autopilot control. :-)
So this is one aspect of my autopilot that I personally think is extremely cool. I can develop and tune the autopilot entirely inside flightgear, and then with minimal changes, migrate the config file over to the actual UAV. And in many cases, the result has worked and converged on the first try (although further tuning is almost always needed.)
-
Permalink Reply by Curt Olson on
-
Probably not too many are interested, but I dug up some old video of my telemaster from the summer of 2007 showing some of it's first flights and some autonomous work using the Xbow MNAV (now discontinued.) At the time I was struggling with an MNAV firmware bug that would cause the servos to randomly go hard over full stop and stay there until the unit was power cycled. The MNAV firmware was thankfully released as open-source so I was able to spend about 3 weeks tearing my hair out and finally found a buffer over flow in it's serial input routines ... and that fixed up the unit so it could fly reliably.
Video posted here:
http://www.flightgear.org/blogs/category/curt/uas/telemaster/
- ‹ Previous
- 1
- 2
- Next ›
Social Networking
Top Content
Contests
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.A list of all T3 contests is here.
Groups
-
Agricultural UAVs
155 members
-
Moderators
24 members
-
ArduRover User Group
205 members
-
ArduPlane User Group
719 members
-
Spanish arducopter Users
88 members
Advertisement
© 2013 Created by Chris Anderson.
Powered by
