[GRASS-user] v.isochrones : error message

Moritz Lennert mlennert at club.worldonline.be
Sun Apr 5 16:15:48 PDT 2015


On 06/04/15 00:02, Moritz Lennert wrote:
> On 05/04/15 12:53, image93 wrote:
>> Hello,
>>
>> Below, the  computational region setting :
>> --------------------------------------------------------------------
>> projection: 99 (Lambert Conformal Conic)
>> zone:       0
>> datum:      towgs84=0,0,0,0,0,0,0
>> ellipsoid:  grs80
>> north:      7037330.05419347
>> south:      -0.94580653
>> west:       0
>> east:       798655
>> nsres:      1
>> ewres:      1
>> rows:       46916
>> cols:       5324
>> cells:      249780784
>> --------------------------------------------------------------------
>>
>> When i modify the resolution nres/ewres from 1 to "150", i have no longer
>> error message. But when i try a process with a resolution with a value <
>> 150, the process is endless...
>>
>> According to you, the results generated at this resolution (150) would be
>> relevant? valid?
>>
>> I loaded the output vector into qgis. The results are seem incoherent.
>> The
>> curves show that we can go in 15 minutes on the other side of Paris! In
>> reality, From my starting point we would go much further toward toward
>> north
>> than south west...
>> I use OSM data. I edited and populated the max speed field thanks to the
>> value of another filed (type of road).
>> How to explain this incoherence? Due to v.isochrones? Due to
>> resolution? Due
>> to input maxspeed values ?
>
> Difficult to say without the data. At this stage, v.isochrones should
> probably be considered beta, notably because of a significant rewrite,
> so it might be v.isochrone's fault. Then again, as I mentioned in my
> response to your other mail, I'm not sure that you are calling
> v.isochrones with the correct input data.
>
> Could you make your data available off-list, so that I can experiment
> with it myself ?

I've done a quick trial with OSM data for Ile-de-France treated as follows:

- reprojected to EPSG 2154 using ogr2ogr
- imported into GRASS

# just to make it quick I just set all roads that do not have a maxspeed 
to maxspeed = 50:
- v.extract roads2154 where="type like 'primary%' OR type like 
'secondary%' OR type like 'tertiary%' OR type like 'motorway%' OR type 
like 'residential%'" output=roads	


- v.db.update roads col=maxspeed value="50" where="maxspeed is NULL"
- v.db.addcolumn roads col="length double precision"
- v.to.db roads op=length col=length
- v.db.addcolumn roads col="cost double precision"
- v.db.update roads col=cost qcol="(length/(maxspeed*1000))*60"
- v.net -c roads op=nodes out=roads_with_nodes
- g.region vect=roads_with_nodes res=10 -ap
- echo "652625|6864473" | v.in.ascii in=- out=start
- v.isochrones roads_with_nodes start=start time_steps=15,30,45 
isochrones=isochrones method=r.cost cost_column=maxspeed
(time = 3m20)

and

- v.isochrones roads_with_nodes start=start time_steps=15,30,45 
isochrones=isochrones method=v.net.iso cost_column=cost max_distance=1000
(time = 1m9)

See attached images for the result (only motorways and primary roads shown).

Moritz
-------------- next part --------------
A non-text attachment was scrubbed...
Name: isochrones_rcost_example.png
Type: image/png
Size: 201201 bytes
Desc: not available
URL: <http://lists.osgeo.org/pipermail/grass-user/attachments/20150406/4371017f/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: isochrones_vnetiso_example.png
Type: image/png
Size: 152672 bytes
Desc: not available
URL: <http://lists.osgeo.org/pipermail/grass-user/attachments/20150406/4371017f/attachment-0003.png>


More information about the grass-user mailing list