1
0
Fork 0
Commit graph

15 commits

Author SHA1 Message Date
curt
7cc98ad15f I tested:
LaRCsim c172 on-ground and in-air starts, reset: all work
UIUC Cessna172 on-ground and in-air starts work as expected, reset
results in an aircraft that is upside down but does not crash FG.   I
don't know what it was like before, so it may well be no change.
JSBSim c172 and X15 in-air starts work fine, resets now work (and are
trimmed), on-ground starts do not -- the c172 ends up on its back.  I
suspect this is no worse than before.

I did not test:
Balloon (the weather code returns nan's for the atmosphere data --this
is in the weather module and apparently is a linux only bug)
ADA (don't know how)
MagicCarpet  (needs work yet)
External (don't know how)

known to be broken:
LaRCsim c172 on-ground starts with a negative terrain altitude (this
happens at KPAO when the scenery is not present).   The FDM inits to
about 50 feet AGL and the model falls to the ground.  It does stay
upright, however, and seems to be fine once it settles out, FWIW.

To do:
--implement set_Model on the bus
--bring Christian's weather data into JSBSim
-- add default method to bus for updating things like the sin and cos of
latitude (for Balloon, MagicCarpet)
-- lots of cleanup

The files:
src/FDM/flight.cxx
src/FDM/flight.hxx
-- all data members now declared protected instead of private.
-- eliminated all but a small set of 'setters', no change to getters.
-- that small set is declared virtual, the default implementation
provided preserves the old behavior
-- all of the vector data members are now initialized.
-- added busdump() method -- FG_LOG's  all the bus data when called,
useful for diagnostics.

src/FDM/ADA.cxx
-- bus data members now directly assigned to

src/FDM/Balloon.cxx
-- bus data members now directly assigned to
-- changed V_equiv_kts to V_calibrated_kts

src/FDM/JSBSim.cxx
src/FDM/JSBSim.hxx
-- bus data members now directly assigned to
-- implemented the FGInterface virtual setters with JSBSim specific
logic
-- changed the static FDMExec to a dynamic fdmex (needed so that the
JSBSim object can be deleted when a model change is called for)
-- implemented constructor and destructor, moved some of the logic
formerly in init() to constructor
-- added logic to bring up FGEngInterface objects and set the RPM and
throttle values.

src/FDM/LaRCsim.cxx
src/FDM/LaRCsim.hxx
-- bus data members now directly assigned to
-- implemented the FGInterface virtual setters with LaRCsim specific
logic, uses LaRCsimIC
-- implemented constructor and destructor, moved some of the logic
formerly in init() to constructor
-- moved default inertias to here from fg_init.cxx
-- eliminated the climb rate calculation.  The equivalent, climb_rate =
-1*vdown, is now in copy_from_LaRCsim().

src/FDM/LaRCsimIC.cxx
src/FDM/LaRCsimIC.hxx
-- similar to FGInitialCondition, this class has all the logic needed to
turn data like Vc and Mach into the more fundamental quantities LaRCsim
needs to initialize.
-- put it in src/FDM since it is a class

src/FDM/MagicCarpet.cxx
 -- bus data members now directly assigned to

src/FDM/Makefile.am
-- adds LaRCsimIC.hxx and cxx

src/FDM/JSBSim/FGAtmosphere.h
src/FDM/JSBSim/FGDefs.h
src/FDM/JSBSim/FGInitialCondition.cpp
src/FDM/JSBSim/FGInitialCondition.h
src/FDM/JSBSim/JSBSim.cpp
-- changes to accomodate the new bus

src/FDM/LaRCsim/atmos_62.h
src/FDM/LaRCsim/ls_geodesy.h
-- surrounded prototypes with #ifdef __cplusplus ... #endif , functions
here are needed in LaRCsimIC

src/FDM/LaRCsim/c172_main.c
src/FDM/LaRCsim/cherokee_aero.c
src/FDM/LaRCsim/ls_aux.c
src/FDM/LaRCsim/ls_constants.h
src/FDM/LaRCsim/ls_geodesy.c
src/FDM/LaRCsim/ls_geodesy.h
src/FDM/LaRCsim/ls_step.c
src/FDM/UIUCModel/uiuc_betaprobe.cpp
-- changed PI to LS_PI, eliminates preprocessor naming conflict with
weather module

src/FDM/LaRCsim/ls_interface.c
src/FDM/LaRCsim/ls_interface.h
-- added function ls_set_model_dt()

src/Main/bfi.cxx
-- eliminated calls that set the NED speeds to body components.  They
are no longer needed and confuse the new bus.

src/Main/fg_init.cxx
-- eliminated calls that just brought the bus data up-to-date (e.g.
set_sin_cos_latitude). or set default values.   The bus now handles the
defaults and updates itself when the setters are called (for LaRCsim and
JSBSim).  A default method for doing this needs to be added to the bus.
-- added fgVelocityInit() to set the speed the user asked for.  Both
JSBSim and LaRCsim can now be initialized using any of:
vc,mach, NED components, UVW components.

src/Main/main.cxx
--eliminated call to fgFDMSetGroundElevation, this data is now 'pulled'
onto the bus every update()

src/Main/options.cxx
src/Main/options.hxx
-- added enum to keep track of the speed requested by the user
-- eliminated calls to set NED velocity properties to body speeds, they
are no longer needed.
-- added options for the NED components.

src/Network/garmin.cxx
src/Network/nmea.cxx
--eliminated calls that just brought the bus data up-to-date (e.g.
set_sin_cos_latitude).  The bus now updates itself when the setters are
called (for LaRCsim and JSBSim).  A default method for doing this needs
to be added to the bus.
-- changed set_V_equiv_kts to set_V_calibrated_kts.  set_V_equiv_kts no
longer exists ( get_V_equiv_kts still does, though)

src/WeatherCM/FGLocalWeatherDatabase.cpp
-- commented out the code to put the weather data on the bus, a
different scheme for this is needed.
2000-10-24 00:34:50 +00:00
curt
df26bde4c4 Updates from Christian Mayer to put some of the weather data "on the bus." 2000-09-30 03:55:57 +00:00
curt
655ce0e20d MACOS -> macintosh 2000-09-10 00:04:50 +00:00
curt
0ffa19cd32 Updates to JSBsim from Jon's CVS.
Massaging some names inside of SimGear.
2000-07-06 22:13:24 +00:00
curt
b074559216 MSVC5 portability changes contributed by Bruce Finney. 2000-05-24 06:10:01 +00:00
curt
519996f34d Changes contributed by Tony Peden to put wind data "on the bus". 2000-05-11 19:17:33 +00:00
curt
485230b443 Code reorganization. 2000-02-15 03:30:01 +00:00
curt
9a207fb4e1 MacOS fixes contributed by Darrell Walisser on 1/26/2000. 2000-02-10 23:37:56 +00:00
curt
210e87ec3a The WeatherDatabase doesn't need the voronoi code anymore but uses
Dave Eberly's spherical interpolation code (found in the Lib/Math
directory). So it would be great if you could give him also a place
in the thanks file.  Changing the WeatherDatabse made actually a heavy
internal redesign necessary but no code outside the database is
affected (isn't code hiding great?).
1999-12-23 16:54:54 +00:00
curt
b430d919d4 Updates by Christian Mayer. 1999-11-03 18:06:52 +00:00
curt
adf5011f80 Fixes to Christian's latest weather submission. 1999-10-21 22:53:12 +00:00
curt
2751545008 Updates by Christian Mayer. 1999-10-20 22:19:07 +00:00
curt
eddea671a2 Upgraded to latest weather data base code from Christian Mayer's and make
it the default.
1999-10-14 20:30:54 +00:00
curt
a0aee5d866 Updates by Christian Mayer. 1999-10-12 03:30:24 +00:00
curt
b645c8cf7e Initial revision. 1999-08-10 03:40:34 +00:00