I don't know how detailed you want to get with the digitrak modeling,
but as a first pass, you could just assume that what ever the digitrak
is doing, it's keeping a pretty good estimate of reality. If you make
that assumption, then you could just use the raw pitch, roll, yaw values
from FG and move on to the higher level modeling. As Paul suggests, if
you want to really model all the internal sensors, you really have to
know exactly what they are, how they are calibrated to some absolute
reference, what math is going on behind the scenes, etc. etc.
Curt.
Paul Kahler wrote:
On Wed, 2005-09-14 at 20:35 -0400, Steve Knoblock wrote:
The Digitrak is described as employing "gyroscopic rate sensors are
installed so as to sense motion about each of the major axes (roll,
yaw and pitch)."
I assume they mean there is a spinning gryo around which three sensors
are arranged, to sense motion in each axis, pitch, roll and yaw. The
sensors report how much the aircraft has moved around the gyro for
each axis.
Gyroscopic rate sensors measure rotation rate around a single axis with
respect to whatever they're mounted on. To measure 3 axis motion
requires a minimum of 3 sensors. There are no gimbals or spinning
objects (vibrating parts yes). Tracking orientation with them requires
integrating the signals in a rotating reference frame. You also need an
absolute reference because the integrated signals will drift. Here's
something google turned up:
http://www.smalltimes.com/document_display.cfm?section_id=76&document_id=7451
From what I can see of the various default instruments in FlightGear,
the only source of roll angle from an instrument is the attitude
indicator or indirectly, the turn coordinator, which the Digitrak does
not use.
I conclude that to model the Digitrak fully, I would need to create C
code to represent this three axis gyro using the gyro.*** code that
the attitude indicator depends on. I have a little experience with C,
but not much. I nearly understand how the attitude indicator works
with the gyro model, but I still have to many questions to comprehend
all it is doing.
I also assume that using /orientation/roll-angle is the best
substitute currently available.
I would appreciate any help with this and please correct me if I am
wrong in any of this.
The only way to really know what the Digitrak is doing is to know what
it's really doing ;-)
I think the Digitrak would make an interesting contribution to
FlightGear.
It would, but you may be limited to just a visual representation and
some other autopilot code. It would be really hard to know exactly what
their software is doing without access to source code.
--
Curtis Olson http://www.flightgear.org/~curt
HumanFIRST Program http://www.humanfirst.umn.edu/
FlightGear Project http://www.flightgear.org
Unique text: 2f585eeea02e2c79d7b1d8c4963bae2d
_______________________________________________
Flightgear-devel mailing list
Flightgear-devel@flightgear.org
http://mail.flightgear.org/mailman/listinfo/flightgear-devel
2f585eeea02e2c79d7b1d8c4963bae2d
