Curt, what is your process for going from this flight data to a JSBSim
model? What sort of maneuvers do you fly and how to get extract the
FDM parameters?
Jeff,
Now you are asking hard questions. I'm not an aerospace engineer so I am trying to learn the process as I go. I know people do hang up a model and swing it to get the moments of inertia. What they measure and how they crunch the results to get the numbers that FlightGear wants ... that I do not know. I know you can fly specific flight tests and extract some specific parameter(s) and then twiddle your sim numbers to get that to match ... for instance, establish straight and level flight at 90 kts (+ some specific alt/press/temp), give a 10 degree aileron deflection, record the roll rate ... now go make your sim results match under those exact same conditions. Repeat for about 100-1000 other tests ... phugoid, pitch rates, flap changes, speed changes, acceleration, deceleration, trim points, control surface deflection speeds, etc. etc.
I begged the gurus over at flightgear-flightmodel to help me get an initial guess at a JSBSim config for my Rascal 110. Then I took that and tweaked it in several ways until certain things felt a lot better ... roll rates, adverse yaw, pitch change with throttle change, and probably a few other things that I noticed really needed to be fixed. But this was all based on zero hard data, and only my recollection of my original perceptions ... not exactly a firm foundation.
Now (well hopefully next week) I'll have an IMU/GPS unit installed in the Rascal 110 that can also record control inputs (from which I can derive things like control surface deflection angles.) So starting next week I should be able to start collecting some decent hard data ... what is the stall speed really? What is my top speed really? What control surface deflections are required to achive straight and level flight at different speeds? What is my decent rate at idle? What's my climb rates, roll rates, pitch rates, etc.? This is probably backwards from the way a smart aerospace person would approach the problem, and hopefully someday I'll learn smart ways to do these things, but for now, this is the approach I have in mind: Fly specific tests, record specific results, twiddle simulator coefficients to try to match the real results.
Hopefully my embedded flight computer will someday be running nasal (no really, actually not joking here) and then I can script out specific flight tests so the computer can fly them with extreme precision, both in real life and in FlightGear and then hopefully be able to make very valid and very direct comparisons between my real flight data and my simulation results.
And hopefully you noticed that much of this message is written in the "future" tense so bear in mind that I could just be full of a lot of hot air. I think I know what I plan to do with this, but I haven't actually done it yet, and I'm sure the real world will have something to say about my level of success, and that might (or will!) dictate at least some change in strategy along the way.
Curt.
--
Curtis Olson - University of Minnesota - FlightGear Project
http://baron.flightgear.org/~curt/ http://www.humanfirst.umn.edu/ http://www.flightgear.org
Unique text: 2f585eeea02e2c79d7b1d8c4963bae2d
------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
