The new multiplayer patch with an extension to transmit some properties with
the base package. The properties are transmitted in a way that will not
immediately brake the packet format if we need new ones.
Even if the maxmimum number needs to be limited somehow, that format might
work well until we have an improoved packet format which is even more compact
and that does not require to retransmit redundant information with each
packet.
That part is relatively fresh and based on that what Oliver provides on his
multiplayer server web page.
The properties are transferred to the client and I have modified the seahawks
rudder animation property to use a relative property path to verify that it
works appart from the fact that you can see it changing in the property
browser.
The movement is still a bit jerky, but that can be fixed/tuned later without
again braking the packet format.
This patch removes some useless indirection when creating AIModels. It
obsolets AIScenario*.
AIEntities are just an intermediate copy of an other intermediate copy of an
xml file on the way from the ai scenario configuration file to the AIModels.
As such the AImodels can now be created directly from the property tree read
from the scenario file.
This reduces the amount of work needed to add an other AIModel and reduces the
amount of copy operations done during initialization.
It also moves internal knowledge of special AI models into these special AI
models class instead of spreading that into the whole AIModel subdirectory
which in turn enables to use carrier internal data structures for carrier
internal data ...
Also some unused variables are removed from the AIModel classes.
I believe that there are still more of them, but that is what I stumbled
accross ...
Tested, like the other splitouts these days in a seperate tree and using the
autopilot for some time, and in this case with a carrier start ...
This patch is a combined effort by Gregor Richards, Oliver Schroeder, and
Vivian Meazza (and code cleanups and improvements by Erik Hofman). It corrects
the bug in which a Multiplayer model responds to local inputs, and the view
number bug which caused certain aircraft to appear as cockpit only models. It
passes remote properties over the net, and all major control surfaces and gear
are now animated correctly, providing that the local ~model.xml file contains
no leading "/" in the <property></property> data entries. MP objects
are now extrapolated using 1st and 2nd derivatives to make their movement
appear more smooth. The sim is now halted while a new client joins the net.
Known problems with MP are non-display of the remote client under certain
circumstances of starting/resetting, and a freeze on starting. These bugs are
long standing, and are not addressed by this patch.
Special thanks must go to AJ Macleod for his patient testing of this patch over many evenings.
We have also moved part of multiplayer into AIModels as part of the ongoing
development of MP.
I had a quick view over the ssgBase::ref() calls in flightgear.
I made them all symmetric and used ssgDeRefDelete to dereference them.
This has the basic advantage that ssgDeRefDelete additionaly deletes the
memory instead of just decrementing the reference cound without deletion ...
This includes an incorrect deref instead of a ssgDeRefDelete in the placement
transform registration I introduced earlier. I believe that this causes the
problems with long flights (unverified, but with a big propability).
There was a patch from Manuel Masing a few months ago which cleaned up
SGLocation's way depending on input values. That means that with that patch
SGLocation does no longer have calls with unneeded input arguments.
I took his patch and integrated that into flightgear and made maximum use of
that changes.
Erik Hofman:
Remove some duplicate code that was moved to simgear/compiler.h
Some quite extensive changes to the AIModel code:
1. Mathias has made major changes to the AICarrier code to provide better
alignment of an aircraft on deck with the carrier - this feature is a major
improvement on the existing, but has a bug which might cause it to fail when
the computer carries out other tasks - changing window size is a known
example. This bug is outwith this code.
2. I have made significant changes to the AIShip code to enable a ship the
turn and roll smoothly.
3. I have added some simple AI which enables the carrier to remain within,
or return to, an operating box.
4. An automated turn into wind for flying operations.
5. A simplistic implementation of TACAN within AICarrier. I am in the course
of implementing this as a generic instrument, but this is some time off
completion.
- AIManager.cxx :
- we can now have multiple <scenario> entries in the sim/ai entry in preferences.xml
- AIBase.cxx :
- added an exception handler around the loading of the 3D model to not exit FG
if the model is not found
- AIScenario.cxx :
- removed a duplicated read of the xml file, this was also exiting FG is the xml file
does not exist
I have introduced the posibility to start directly on the carrier.
With that patch you will have a --carrrier=id argument where id can either be
the pennant number configured in the nimitz scenario or the carriers name
also configured in the carriers scenario.
Additionaly you can use --parkpos=id to select different positions on the
carrier. They are also configured in the scenario file.
That includes the switch of the whole FGInterface class to make use of the
groundcache.
That means that an aircraft no longer uses the current elevation value from
the scenery class. It rather has its own local cache of the aircrafts
environment which is setup in the common_init method of FGInterface and
updated either manually by calling
FGInterface::get_groundlevel_m(lat, lon, alt_m);
or implicitly by calling the above method in the
FGInterface::_updateGeo*Position(lat, lon, alt);
methods.
A call get_groundlevel_m rebuilds the groundcache if the request is outside
the range of the cache.
Note that for the real usage of the groundcache including the correct
information about the movement of objects and the velocity information, you
still need to set up the groundcache in the usual way like YASim and JSBSim
currently does.
If you use the native interface, you will get only static objects correctly.
But for FDM's only using one single ground level for a whole step this is IMO
sufficient.
The AIManager gets a way to return the location of a object which is placed
wrt an AI Object. At the moment it only honours AICarriers for that.
That method is a static one, which loads the scenario file for that reason and
throws it away afterwards. This looked like the aprioriate way, because the
AIManager is initialized much later in flightgears bootstrap, and I did not
find an easy way to reorder that for my needs. Since this additional load is
very small and does only happen if such a relative location is required, I
think that this is ok.
Note that moving on the carrier will only work correctly for JSBSim and YASim,
but you should now be able to start and move on every not itself moving
object with any FDM.
I added an AIStatic object to my OV-10 sim for use in putting city signs,
vehicles, or anything else that will be static, but that I don't want to put
in the scenery files. It's inexpensive. Before, I was making such things
from AIShip.
I also added the ability to set flight plans to repeat, so that when an
airplane reaches the end it just starts over at the beginning. This is
useful for my OV-10 sim. I have C-141 and KC-135 traffic flying approaches
to Ramstein, and I only have to define two AI objects to do this.
Also, I found an inefficiency in AIBase, where every AI object was calculating
Mach number at every dt. Now only AIBallistic objects do this.
I have done a patch to eliminate the jitter of 3D-objects near the viewpoint
(for example 3D cockpit objects).
The problem is the roundoff accuracy of the float values used in the
scenegraph together with the transforms of the eyepoint relative to the
scenery center.
The solution will be to move the scenery center near the view point.
This way floats relative accuracy is enough to show a stable picture.
To get that right I have introduced a transform node for the scenegraph which
is responsible for that shift and uses double values as long as possible.
The scenery subsystem now has a list of all those transforms required to place
objects in the world and will tell all those transforms that the scenery
center has changed when the set_scenery_center() of the scenery subsystem is
called.
The problem was not solvable by SGModelPlacement and SGLocation, since not all
objects, especially the scenery, are placed using these classes.
The first approach was to have the scenery center exactly at the eyepoint.
This works well for the cockpit.
But then the ground jitters a bit below the aircraft. With our default views
you can't see that, but that F-18 has a camera view below the left engine
intake with the nose gear and the ground in its field of view, here I could
see that.
Having the scenery center constant will still have this roundoff problems, but
like it is now too, the roundoff error here is exactly the same in each
frame, so you will not notice any jitter.
The real solution is now to keep the scenery center constant as long as it is
in a ball of 30m radius around the view point. If the scenery center is
outside this ball, just put it at the view point.
As a sideeffect of now beeing able to switch the scenery center in the whole
scenegraph with one function call, I was able to remove a one half of a
problem when switching views, where the scenery center was far off for one or
two frames past switching from one view to the next. Also included is a fix
to the other half of this problem, where the view position was not yet copied
into a view when it is switched (at least under glut). This was responsible
for the 'Error: ...' messages of the cloud subsystem when views were
switched.
Okay, here's the latest update to the tarffic manager/AI Manager. AITraffic
can now fly multiple routes and be initialized while sitting statically at
airports.
The moving ai models will jump around realtive to the moving aircraft model.
I can see that with the carrier but others have noticed that too with ai
aircraft before.
The reason is that all SGSystems are called with a dt value which is not
necessarily a multiple of 1/hz.
In contrast, most FDM's use the _calc_multiloop function from FGInterface
which forces the time update to be a multiple of 1/hz for the FDM aircraft.
As a result, in the worst case, the FDM aircraft has moved nearly 1/hz seconds
further than the rest of flightgear (1/120sec*300kts that is about 1.3m).
That patch forces the time update to be a multiple of 1/hz.
The value of rho (air density) varies with height. (Including the upper
stratosphere, ust in case someone wants to model ICBMs.) The standard
atmosphere is used (based on a sea-level temperature of 15 deg C.).
Erik Hofman:
I moved this code over the AIBase::update() so all AIModels can make
use of rho, temperature, pressure, etc.
I have added <Cd> and <weight> to the input parameters in the submodels.xml
script. Raw data may be used, thus avoiding the need to guestimate <eda>.
Eda remains, but should now be used to enter the proper cross-sectional
area.
Here's some new AI stuff.
1) AI objects must now be defined in a scenario file, not in preferences.xml
or a *-set file. (Of course this doesn't prevent objects from being created
dynamically, as with Durk's traffic manager).
2) A new demo_scenario file is attached. It creates 3 aircraft, a sailboat,
and a thunderstorm.
3) Objects without flightplans live forever.
4) FGAIShip::ProcessFlightplan() is not yet implemented.
5) preferences.xml should now define only <enabled> and <scenario>
1. Removed aircraft roll on ground.
2. Decreased descent pitch angle.
3. Updated flightplans to include <on-ground>
4. Fixed property indexing, so all AI aircraft have their own property branch
The default value of <on-ground> is false, so you only need to specify it when
on the ground. For takeoff you need to specify <on-ground>true</on-ground>
for the first waypoint, and for the acceleration waypoint. For landing you
need to specify it for the touchdown point and any taxi points.
One problem. WARNING **** There is a bug in the way the property system
works, which causes a segfault, but I don't know if the problem is in the
property code, or in how I'm using it. After an AI object terminates, if you
access the property tree through the property browser the sim will segfault.
Here's the newest AI stuff.
The AIManager at init() creates a new scenario. Right now the
default_scenario is hard coded in, but eventually the AIManager should get
the scenario filename from preferences.xml.
The scenario defines which AI objects will be created. Right now it only
creates AIAircraft, but this is easily extended. The scenario also defines
which flightplan will be assigned to the airplane. Scenario config files go
in data/Data/AI.
The Airplane gets a pointer to a FlightPlan object. Each airplane should get
its own flightplan object, even if two airplanes have the same flight plan.
This is because the flightplan maintains the iterator pointing to the
current waypoint, and two airplanes might be at different locations (for
instance if they were created at different times). The flight plan files go
in data/Data/AI/FlightPlans.
When the airplane gets to the waypoint named "END" it vanishes. The
AIAircraft destructor deletes its flight plan (if it has one).
The last waypoint is a place holder only. I called mine
<WPT><NAME>"EOF"</NAME></WPT>.
I added some things to the AI stuff to improve the AIThermal processing.
Before, all the thermals were processed in order, and the last one overwrote
the prior one. Now, only the data from the nearest thermal is kept. This
way a tile can be populated with many thermals, and (as long as they have the
same diameter) the one nearest the airplane correctly takes effect. This
will make us ready for the next step, "auto-thermaling", where FlightGear's
tile manager can cover a tile with thermals, and set the thermal strength
based on land-use type.
I moved the enumerated object_type to the base class. When an AI object is
created it now sets the _otype variable in the base class. This lets the AI
manager find out what kind of AI object it is dealing with, using the base
pointer. I also added a function isa() to the base class, so the manager can
process objects differently based on their type.
The AI manager now sends AIThermal processing to a different function, where
only the data from the nearest thermal is kept. After the manager processes
all the AI objects, then the results from the nearest thermal are applied to
wind-from-down.
Here's a new batch of AI code which includes a working radar instrument.
I put the radar calculations into the existing AIAircraft class. It was
easier that way, and it can always be migrated out later if we have to.
Every tenth sim cycle the AIManager makes a copy of the current user state
information. When the AIAircraft updates it uses this information to
calculate the radar numbers. It calculates:
1) bearing from user to target
2) range to target in nautical miles
3) "horizontal offset" to target. This is the angle from the nose to the
target, in degrees, from -180 to 180. This will be useful later for a HUD.
4) elevation, in degrees (vertical angle from user's position to target
position)
5) vertical offset, in degrees (this is elevation corrected for user's pitch)
6) rdot (range rate in knots, note: not working yet, so I commented it out)
and three items used by the radar instrument to place the "blip"
7) y_shift, in nautical miles
8) x_shift, in nautical miles
9) rotation, in degrees
The radar instrument uses the above three items, and applies a scale factor to
the x-shift and y-shift in order to match the instrument's scale. Changing
the display scale can be done entirely in the XML code for the instrument.
Right now it's set up only to display a 40 mile scale.
The radar is an AWACS view, which is not very realistic, but it is useful and
demonstrates the technology. With just a little more work I can get a HUD
marker. All I need to do there is make a bank angle adjustment to the
current values.
I went through the AI code to put the "bank" node back into the config file,
so the models can fly circles. While I was in there I made some other
changes.
*) Moved the initialization of roll, tgt-roll, pitch ... etc, from init()
into the constructor, so it wouldn't over-write the config settings.
*) Changed the altitude getter to remove the meters-to-feet conversion. The
altitude is kept internally in feet. Only the scenery code needs meters.
*) Added "bank" item for config file (for type=aircraft). Left bank is
negative.
*) Added "rudder" item for config file (for type=ship). Left rudder is
negative. Internally this is stored in the "roll" variable, but the ship
model doesn't roll. It uses the "roll" variable for turning though.
The following puts a tanker at 3000 feet, 6 nm northwest of KSFO. On takeoff,
the tanker is visible over the hanger building at one-o'clock.
<entry>
<type>aircraft</type>
<class>jet_transport</class>
<path>Aircraft/737/Models/boeing733.xml</path>
<speed-KTAS type="double">320.0</speed-KTAS>
<altitude-ft type="double">3000.0</altitude-ft>
<longitude type="double">-122.455</longitude>
<latitude type="double">37.69667</latitude>
<heading type="double">200.0</heading>
<bank type="double">-15.0</bank>
</entry>
