David Megginson writes:
> It might be worth finishing the long-delayed overhaul of the FDM
> interface first. Andy demonstrated initially how effectively an FDM
> can work almost exclusively through the property manager, and most new
> JSBSim functionality has bypassed FGInterface and gone directly
> through the property manager as well. Basically, FGInterface just
> needs to be a regular subsystem:
>
> class FGInterface : public FGSubsystem
> {
>
> public:
> FGInterface ();
> virtual ~FGInterface ();
> virtual void init ();
> virtual void update (double dt_sec);
> virtual void bind ();
> virtual void unbind ();
>
> };
>
> We could stick some convenience property node pointers in there
>
> protected:
> SGPropertyNode_ptr longitude_deg_node;
> SGPropertyNode_ptr latitude_deg_node;
> SGPropertyNode_ptr altitude_ft_node;
> SGPropertyNode_ptr roll_deg_node;
> SGPropertyNode_ptr pitch_deg_node;
> SGPropertyNode_ptr heading_deg_node;
>
> and so one for the most basic stuff that applies to all models
> (velocities and environment are also good bets), but we wouldn't want
> to get carried away.
>
> Comments? This should make integrating any flight model fairly
> straight forward -- it implements and init() call and an update()
> call, where the update gives the delta time in seconds since the last
> call. Any information it needs is available through the property tree
> (i.e. /controls/collective). It makes for a very shallow learning
> curve.
I'm onboard I think, although I don't have time to work on this
myself.
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