I wrote:

> Sent: 19 December 2007 09:17
> To: 'FlightGear developers discussions'
> Subject: RE: [Flightgear-devel] External Cargo, was: Re: 
> screenshots (and "snapshots")
> 
> 
> 
> 
> > -----Original Message-----
> > From: [EMAIL PROTECTED]
> > [mailto:[EMAIL PROTECTED] On 
> > Behalf Of Maik Justus
> > Sent: 18 December 2007 23:12
> > To: FlightGear developers discussions
> > Subject: Re: [Flightgear-devel] External Cargo, was: Re: 
> > screenshots (and "snapshots")
> > 
> > 
> > Hi Viviam,
> > Vivian Meazza schrieb am 18.12.2007 23:41:
> > > I've just a few moments ago added a few lines of code to
> > AIBallistic
> > > which apply an external force (using the same math as
> > above). The rest
> > > works as before. I just need few properties to set the
> > magnitude and
> > > vector of the external force and it will be done, at least for
> > > testing.
> > Thank you, very good. Is it possible to pass the forces in the earth
> > fixed coordinates axes base? (in Newton or pof)?
> 
> I'm planning, and have tested the passing of the force in 
> terms of magnitude (pound force), azimuth (degrees, north = 
> 0) and elevation (degrees, horizontal = 0, up = 90). I hope 
> this is what you need, and can work with.
> 
> > > It's trivial
> > > really, but I'm assuming that the external force applies no
> > rotational
> > > force to the object.
> > I think this could be easily faked. e.g. new_roll = old_roll +
> > cos(old_roll) * force_in_y_direction * a_magic_constant * dt 
> > but i f I read the code correctly, yaw and hdg are pointing 
> > at the speed 
> > direction (slightly filtered)? Therefore we need something 
> > like _elevation = atan2(force_in_z_direction,force_in_x_direction)
> > and then your filter to add some inertia?
> 
> Right now there are 2 options - 1, the ballistic object 
> aligns with the velocity vector (damped), or 2, it remains 
> static. These options both work with the new code. I think 
> option 1 will do for an under slung load?
>  
> > > I haven't a great deal of time available in the run up to
> > Christmas so
> > > no promises on completion.
> > >   
> > Same here. I will have very limited access to my computer (family
> > visiting tour). Fortunately my computer is a notebook, but I 
> > am not sure 
> > if there is any space in the car for my joystick.
> 
> We have a horde of family visitors over the coming days - all 
> competing for my time and the computer. We'll see how much 
> can be done (not a lot I expect).
> 

Christmas has arrived slightly early! I've got something which:

A. runs

B. looks OK with limited testing

The ballistic object aligns with the direction of flight in pitch and
heading with an external force applied. It would be possible to align it
with the direction of the external force, but I think that would need roll
as well. I'm not sure which one would look best.

The external force is defined in terms of:

Magnitude (lbf)
Azimuth (deg, North = O)
Elevation (deg, up = 90)

In a user-defined property. Of course, some external program needs to set
the external force data.

This all now needs testing in a more realistic environment. I'm not totally
convinced that the ballistic object won't disappear into space/to the centre
of the earth, or oscillate like a deranged spring.

Vivian

  



-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel

Reply via email to