On Mon, 2004-04-05 at 16:04, Jim Wilson wrote: > Jon S Berndt said: > > > On Mon, 5 Apr 2004 23:15:42 +0200 > > Mathias Fröhlich <[EMAIL PROTECTED]> wrote: > > > > > >The onground property is now ok. > > >You can reset now JSBSim aircraft. > > >Thanks for the fix! > > > > > > > ?? > > > > Was it "bad data"? > > > > All that was bad was a flag caused the trim routine to be called or not called > with a "tGround" parameter...whatever that is, which was discussed in this > thread yesterday. Questions are still unanswered, so I'm sure that sometime > down the road the trim routine will rear it's ugly head again.
The trimming routine has several modes, one of which is tGround. That tells it to iterate on altitude and pitch attitude until the gear forces balance the weight. tLongitudinal tells it to trim in-air with angle of attack for lift, elevator for pitch, and thrust for speed. tFull tells it to trim all six axes, those above plus adjust the controls, bank angle, and sideslip to trim yaw, roll, and side force. tTurn sets up the body axis rates to perform a tFull trim in a steady state turn. If I knew of a way to make the trimming routine less sensitive to bad data (such as a bad terrain altitude) or divine how it's supposed to set itself up, I'd do it. I can think of things that could be done but they all involve either a sacrifice in capability or attempting to make the trimming routine smarter than it should be. > > David's explanation is relevant but not entirely, because this issue was > showing up doing a full reset. > > I think, that maybe this could be resolved by doing as Andy described earlier. > In other words know where the gear is and get it above the pavement (ground > elevation) before the simulation starts. Then and only then drop the sucker, > start the simulation, and let it settle. Sorry for the simplistic explanation > and my apologies for suggesting that JSBSim could do something about bad data > instead of just crashing random aircraft. Thank you. > As it is now we need to test every > single JSBSim aircraft every time a modification is made to flightgear because > the trim routine is lacks robustness. It does demand alot of both the calling program and the aircraft configs. As much as I'd hate to see it stop getting used, maybe the right thing is too just stop using it altogether. > > Best, > > Jim > > > _______________________________________________ > Flightgear-devel mailing list > [EMAIL PROTECTED] > http://mail.flightgear.org/mailman/listinfo/flightgear-devel -- Tony Peden <[EMAIL PROTECTED]> _______________________________________________ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
