Re: [Autopilot] new to the list

[Tramm H. <hu...@sw...>]

Hi, Jim.

> Hi, all.  I'm Jim Cunningham.  My research interest is in the flight and
> biomechanics of  late-Cretaceous azhdarchid pterodactyls.  I'm currently
> working on the early stages of a full scale flapping replica of
> Quetzalcoatlus species (similar to more widely recognised Quetzalcoatlus
> northropi, but with a little less than half the wingspan) in an effort
> to test some of my computer simulations of the flight kinematics of the
> living animals. The Qsp replica will mass about 20 Kg, with a wingspan
> of approximately 4.8 meters when in soaring configuration.  Aspect ratio
> is about 16.5.

That is quite a creature!  I've read about several ornithopter
projects that are having trouble with thrust to mass ratios.
They work great for small scale models, but as the airframe
scales up the efficiencies start to fail.

Have you had any problems with lifting 20 kg?

> I'd intended to do an autopilot somewhat similar to the
> one Paul MacCready's team used in their standoff half scale 5.5 meter
> span QN replica a few years back,

The Gossamer Condor guy, right?  And it seems he's now doing the
Helios project, too.

> [...] May I
> lurk for a while and attempt to learn more about your autopilot
> techniques?

Of course!  The web site is updated fairly frequently with developments
and plans, most recently related to the IMU and Kalman filtering.  And
you can see the up-to-the-minute source code, too.

I'd be very interested in seeing how an INS would cope with the non-
linear forces that an ornithopter exerts as it climbs.  The periodic
pulses from the wing flaps would seem to contradict many of the
assumptions of smooth, continuous forces.

Trammell
-- 
  -----|----- hu...@sw...                         H 240-476-1373
*>=====[]L\   Tra...@ce...              W 240-453-3317
'     -'-`-   http://www.swcp.com/~hudson/                    KC5RNF