[talk-au] Uploading traces (Was; Hi all ...)

James Livingston doctau at mac.com
Thu Jun 18 10:32:12 BST 2009


I've actually been playing around with some automated-analysis stuff  
recently. In general you need a fairly large amount of traces covering  
a particular area to get anything meaningful out, but it's still  
interesting.

I think that not uploading your gps tracks is time, and there are  
perfectly valid reasons for not doing so. That said, having many gps  
tracks uploaded is actually useful.

On 18/06/2009, at 3:25 PM, Ross Scanlon wrote:
> If your connected to the internet that's fine but it's no use for on  
> the
> road re-routing, unless you have all the gps traces downloaded to your
> gps.
>
> This should be tagged by maxspeed anyway.

Sure, but I'd guess that well over 99% of the roads in Australia don't  
have maxspeed tagged. If we can do some analysis on uploaded GPX  
tracks, and get an addition 0.1% of roads marked with a maxspeed,  
wouldn't that be better than nothing?


> Some one may have also driven the road really slowly (push bike) and  
> some
> one may have done it at the speed limit.  This would skew any  
> reliability.

If you simply take the mean of the tracks in the area, then yes.

What you'd actually do is figure out the distribution of speeds and  
see that in that case there might be a couple of humps, one for  
pedestrians, one for cyclists and one for motor-powered vehicles.  
After that you'd need to take into account that people slow down for  
corners, lots of people speed, speed limits are generally are  
generally in multiple of 10km/hr and not over 110.

It's going to be pretty complex, but if you can get some "free" data  
which is mostly-accurate for a small fraction of the roads, why not?


>> * improving height maps, by taking (lots of) samples where altitude
>> information was present.
>
> Pointless, vertical data is grossly out from a gps you are better off
> using the NASA dem data.

I agree here.


>> * automatically guessing the number of lanes on a road, by looking at
>> the variance of traces over sections in each direction.
>
> Should be tagged anyway (when more than 1) and how do you know it's  
> not an
> accuracy problem.

This is the other (along with maxspeed) thing I've been playing around  
with. You need a lot of tracks to do it. A *lot*.

If gps track uploads were all marked with the type of vehicle (or  
walking) then you could also do things like detecting where cycle  
lanes were.


> We have to trust that osm's are putting in accurate data but from what
> I've seen the data already there is miles better than google maps
> particularly in rural Australia.

I think in general, the more raw data that we have and can potentially  
use if we want to, the better.


-- 
James





More information about the Talk-au mailing list