Re: [Flightgear-devel] Detailed Simulation of Wing Forces

[Curtis O. <cur...@gm...>]

On Tue, Jul 21, 2009 at 12:17 PM, Christopher G. Prince
<ch...@cp...>wrote:

> Hi,
>
> We're working on a project to feed back information about airflow on
> a wing to the pilots skin (tactile feedback). Our project web page is
> here-- http://www.d.umn.edu/~cprince/PubRes/FbF/<http://www.d.umn.edu/%7Ecprince/PubRes/FbF/>
> Right now, we're pretty early in progress and working with simulated
> wings. I've started using XPlane to provide a simulation of wing
> airflow sensors-- by tapping into its simulation of lifting forces on
> the aircraft wing. Can Flight Gear do something similar? Can I get an
> array of data on lifting forces over the wing back from the
> simulator, in real time?
>
> (I've looked through the flight gear manual some, and through the
> flight model docs some and can't see what I'm looking for).


Hi Chris,

FlightGear includes a couple different approaches to modeling the physics of
flight, but I don't think anything we have closely matches the x-plane
approach.  (Every approach has it's strengths and weaknesses, and the
X-Plane approach is interesting as a very coarse grained virtual wind tunnel
that can predict some things about how an airframe might fly, and some of
the data that comes out of this approach could be interesting ... especially
in your case.  That said, the X-plane approach only takes you a certain
distance down the path, and if you are trying to match the known performance
numbers of a known airframe, you might have to learn a little voodoo to make
that happen in x-plane.)

I glanced through some of the information on your site and I like this idea
of exploring tactile feedback for flying.  In a past life I worked for with
a human factors research group at UMN Twin Cities campus and we did some
experiments with providing tactile feed back as a warning system for when
you are drifting out of your lane, or approaching an upcoming corner too
quickly.  In our case we had a vibration system in the seat that could
vibrate the left or right sides independently (for a lane warning) or both
together for a "too fast" warning.  We've also experimented with having the
steering wheel give some force to push you back towards the center of your
lane, and also with having the accelerator pedal push back on your foot if
you are getting too close to the car ahead.  So using tactile feedback like
this as a warning system makes sense to me.

Providing tactile feedback for the feeling of flight is more nebulous.  I'm
still trying to think that all through.  It sounds like you will be using
pressure sensors on the wing, so you need to figure out a system in the
simulator that simulates the pressure that the sensors would see, not so
much the forces.  I have no idea if that's possible in FlightGear, but maybe
some of our flight dynamics engineers will chime in if they have any ideas.

If you want to measure force, then perhaps you could mount some
accelerometers out on the wings?  That might be easier to simulate (or
estimate) in software.  I know that JSBSim has the capability of reporting
accelerations at some offset location on the airframe, but I don't know if
it supports more than one offset accelerometer?

I'm still trying to imagine how the tactile feedback will feel and how that
will conceptually tie in with what the aircraft is doing.  Would you try to
make the air pressure on your individual arms match the air pressure on the
wings?  I suppose a pilot could get his head around that and learn what
different phase of flight feel like.

Good luck on your project, it sounds like fun! :-)

Curt.
-- 
Curtis Olson: http://baron.flightgear.org/~curt/<http://baron.flightgear.org/%7Ecurt/>