Re: [Flightgear-devel] Today's CVS

[Durk T. <d.t...@xs...>]

On Saturday 07 July 2007 11:16, Thomas F=F6rster wrote:
> Am Samstag 07 Juli 2007 09:41 schrieb Vivian Meazza:
> > OK so it's 2 features, not 1 bug - excellent. I also noted that the rou=
te
> > taken from parking to active runway seemed a bit odd, but then I compar=
ed
> > our taxiways to those on Google Earth - we seem to have bits missing
> > alongside 28L, which explains that. Not good for our default airport
> > though.
>
> The route finding is extremely simple. It just calculates the shortest way
> between 2 points (in terms of travel distance) of the ground network. The
> ground network is a directed graph though. Probably there are some
> invisible "one-ways" in it.

I just checked using taxidraw. There are some taxiroutes missing, thus=20
preventing traffic to follow the shortest route. I'm working on it.

> It might be possible to make the way finding more clever (e.g. penalize
> runway crossings etc.), but this would need a more sophisticated groundne=
t,
> with more node types (i.e. holding positions) than now. Which might be a
> good idea anyway.
>
Eventually, that would be the plan. Before we can do this, we need to have=
=20
more sophistcation in the ground network, and consequently, also in it's=20
principal design tool; taxidraw.

Part of that is going to be my summer / fall project.

Cheers,
Durk