On Wed, 2002-01-16 at 17:41, Andy Ross wrote: > Tony Peden wrote: > > Andy Ross wrote: > > > This would be a good feature to look at breaking out of the FDM. At > > > its most reductionist, a FCS system compares (1) pilot control > > > inputs and (2) FDM output to produce control surface positions. > > > None of that requires access to the internals of the FDM. A > > > well-designed FCS would work with any conceivable FDM. > > > > For 90% of what FG is used for this is probably true. For the general > > case, however, it depends on how much data you want to put into the > > property tree and/or on the bus. Even moderately sophisticated flight > > control systems can be dependent on quite a number of parameters. > > Can you think of any specific examples?
How about hinge moments? They'll need cbar and possibly span, dynamic pressure, tail angle of attack (for which downwash is needed). The list is probably longer for some tail-off models. An airplane with a yaw damper and some way of measuring beta might use betadot in the control law. This is the general case, I'm talking about here, so failure modes are fair game as well. What about a hydraulic system failures in a system where different control surfaces are driven by different hydraulic systems? JSBSim can't do all of this now, but it doesn't need to at the moment either. It does, however, have to hold onto it's FCS to preserve even the potential to do a complete model. And, like I said, for 90%, maybe even 99%, of what FG does, what your proposing will probably work. Really, there's not much for > an FCS to measure beyond control inputs and physical state (position, > velocity, acceleration and their rotational analogs). Would that real airplanes were so simple ... >Sure, some > FCS's have very complex *internal* state (Kalman filtering, predictive > algorithms, etc...), but that's, well, internal. It doesn't affect > the FCS's interface to the outside world. > > > Again, viable for 90% of what FG does, but not for the general case. > > Any FCS that incorporates stability augmentation, for example, will > > be very specific to that aircraft (and sometimes even sub-models). > > Sure. I'm not saying that you'd expect the F-16 FCS to work with a > 172 (in fact, I said exactly the opposite). But you *would* expect > and require that the FCS work for both the JSB F-16 and the YASim > F-16. > > In fact, this is a good example: a "real" F-16A (Dunno about the C) > flight control computer takes its input from a set of gyros and from > the position of the stick, and that's it. You can get all of > information that from the property system or FGInterface today, and in > fact I'd be willing to bet that with a little work, you could actually > plug a real-world FCS into a (hypothetical) FlightGear F-16 and get it > to fly the plane appropriately. > > The issue isn't whether the FCS needs to know specific per-aircraft > information. Of course it does. The issue is whether the FCS *code* > requires access to the internals of the FDM. I argue that it does > not. > > Andy > > -- > Andrew J. Ross NextBus Information Systems > Senior Software Engineer Emeryville, CA > [EMAIL PROTECTED] http://www.nextbus.com > "Men go crazy in conflagrations. They only get better one by one." > - Sting (misquoted) > > > _______________________________________________ > Flightgear-devel mailing list > [EMAIL PROTECTED] > http://mail.flightgear.org/mailman/listinfo/flightgear-devel -- Tony Peden [EMAIL PROTECTED] We all know Linux is great ... it does infinite loops in 5 seconds. -- attributed to Linus Torvalds _______________________________________________ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
