Jim Wilson writes:
> Michael Selig <[EMAIL PROTECTED]> said:
>
> > I speculate that the new problem running w/ the MagicCarpet FDM (buried
> > aircraft) is related to the same problem running w/ the LaRCsim/UIUC FDM
> > that I have reported. All the while, this has not been a problem running
> > w/ JSBSim FDM. The problem started around the time that work on the viewer
> > code began.
> >
>
> Err..umm..foul! ;-) If you look at the outputs in "/position/altitude-ft" you
> will note disagreement about where the ground is at starting KSFO.
>
> JSBSim seems to think KSFO is a 0 elevation by the way. If it is, then Yasim
> is wrong, since it reports a different value for alt and alt-agl on startup at
> KSFO. The other FDM's have a completely different take on the situation
> reporting the elevation a foot or two lower than JSBSim and Yasim.
>
> Basically it isn't that much of a problem if FDM's disagree, but the aircraft
> model offsets will need to be set for any discrepency from the default FDM if
> you use another FDM.
Ok, now isn't this very interesting. There is definitely some
confusion going on here. I agree with Jim, it's not the fault of the
viewer code.
Ignoring the issue of the CG possibly moving during the simulation,
the assumption has always been that the FDM returns the position and
elevation of the CG of the vehicle. The viewer (for better or worse)
has always drawn the view point relative to the CG of the vehicle.
Now we have the ability to offset the pilot view point relative to the
CG of the aircraft and have been doing that, but I'm not sure how that
is now being handled in light of the recent view manager overhaul.
I'm not sure where the code now resides, and I'm not sure where the
pilot view point offset is now configured. Jim can probably speak to
that.
So on to the test results. I started up FlightGear with 4 different
FDM's, opened the property browser and examined the
altitude-above-ground-level (AGL) and the altitude-above-sea-level
(ASL). There is definite some loose interpretation going on by the
FDM's that probably should be corrected and made consistant. Let me
give you the conclusion first: Andy, I think YASim has a bug. :-)
JSBSim: 4.428218 AGL
4.427779 ASL
CG is 4.4' above ground level, and 4.4' above sea level
meaning that the ground is at about 0.0' ASL.
YASim: 0.000000 AGL
4.946215 ASL
YASim is saying that the CG is exactly on the ground level
which is physically impossible. It compensates by saying the
ground level is 4.9' above the real ground level. I think
this is a bug, but it probably looks ok visually.
UFO: 0.000000 AGL
-0.000305 ASL
The UFO fdm is putting the CG exactly above ground level and
isn't allowing for any 'landing gear' or thickness of the
vehicle. It appears to be calculating height of the ground
correctly.
LaRCsim: 3.356987 AGL
3.356550 ASL
LaRCsim put's the CG at a different height above ground when at
rest on the ground, but this would probably correspond to
different aircraft geometry which would be normal.
So, I think the answer is that the FDM should report the actual
position and altitude of the 'reference point' whether it be CG or
some other fixed point on the aircraft, it doesn't matter. Somewhere
(Jim W?) we should be able to configure the offset to the pilot view
point, and the viewer code should honor this. (And Andy should fix
the bug in YASim.) :-)
Then I think everyone will be happy. :-)
Regards,
Curt.
--
Curtis Olson IVLab / HumanFIRST Program FlightGear Project
Twin Cities [EMAIL PROTECTED] [EMAIL PROTECTED]
Minnesota http://www.menet.umn.edu/~curt http://www.flightgear.org
_______________________________________________
Flightgear-devel mailing list
[EMAIL PROTECTED]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel
