On Thursday 09 December 2010 11:08:39 Hal V. Engel wrote:
> On Thursday, December 09, 2010 09:11:39 am Curtis Olson wrote:
> > On Thu, Dec 9, 2010 at 10:10 AM, Hal V. Engel <hven...@gmail.com> wrote:
> > > I am going to try using the autopilot between spins and during climb
> > > outs to
> > > get things in a know state (wings level and so on) as things progress
> > > through
> > > the flight.  This should allow for the state of the flight and the
> > > control inputs to sync up enough to get a reliable playback.
> >
> > Hi Hal,
> >
> > This was going to be my suggestion: use various combinations of autopilot
> > modes to control the flight under a variety of conditions.
> >
> > You may find that you need to tune some of the autopilot modes better
> > than what they are by default.
>
> I have already tuned the autopilot some time ago since the default
> autopilot was very unstable in this aircraft.  The real aircraft does not
> have an autopilot so it is not exposed in the cockpit.  I use it primarily
> as an aid for performance testing (IE. climb rates, time to altitude, speed
> tests...).
>
> > You may find that you need to create some new specialized autopilot
> > modes.
>
> For this probably not.
>
> > Armed with some well tuned autopilot modes, you can write a nasal script
> > (or even an external perl/python/ruby/etc. script) that can monitor the
> > flight progress, select and activate the various modes at the correct
> > time, and then also mix in some open-loop control in key places as well.
>
> I don't think I need an automated script.  I mainly need a way to setup the
> aircraft in a known state (wings level, sufficient altitude, know
> airspeed...) at the start of the open-loop control sequences.  I think I
> can do this by hand by stopping the autopilot just before the open-loop
> sequences while recoding the flight data.  The person requesting this is
> only interested in the open loop sequences (IE. stall, spin and recovery). 
> I would like to provide him with a flight that has both a normal spin
> (entered with power off) and a flat spin (entered with a higher power
> setting) sequences.
>
> > I've had some moderately good success doing things like auto take off,
> > auto landing (even in cross winds), scheduling flaps up/down, gear
> > up/down, etc. at the proper time, even flying a standard visual pattern.
> >
> > Nasal gives you all the trig functions, it gives you the ability to save
> > wgs84 coordinates and then compute course/distance between them, or start
> > with a coordinate and project out a new coordinate at some distance and
> > heading.  You have access to all the wind conditions, ground track, true
> > airspeed, orientation, position, velocity, etc.  You can manipulate any
> > and all the controls of the aircraft.  You can even automatically select
> > and manipulate views.
> >
> > Combine the core capabilities of nasal, the "sensor" data available in
> > the flightgear property tree, the sophisticated multi-stage configurable
> > autopilot system, the ability to manipulate anything and everything, and
> > a visit to the aviation formulary web site and you can be pretty
> > dangerous!
> >
> > Take a simple wind triangle formula combined with knowledge of true
> > airspeed and ground track/speed + some well tuned autopilot stages and
> > you can hit your runway touch down point within a cm or two every time
> > ... even with significant winds.
> >
> > With a little effort, it's possible to create some neat fully automated
> > demonstrations.
> >
> > Curt.
>

It would be really, really cool if you could do this using the JSBSim internal 
features so it could run in stand-alone, too.  :)

Ron

------------------------------------------------------------------------------
Oracle to DB2 Conversion Guide: Learn learn about native support for PL/SQL,
new data types, scalar functions, improved concurrency, built-in packages, 
OCI, SQL*Plus, data movement tools, best practices and more.
http://p.sf.net/sfu/oracle-sfdev2dev 
_______________________________________________
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel

Reply via email to