[mkgmap-dev] Curvy routing support: new function?

Thu Jul 3 21:32:27 BST 2014

Hi WanMil,

great work!
A quick test shows that sometimes negative values for curviness() are appearing.
As there is the abs() statement, there should be no negative values
Please find the result attached.

Almost straight roads show a high curviness value. There must be some bug. Unfortunately I'm lost in finding the right position in the source code, so I rely on you :-)

Cheers
Manfred

> Gesendet: Donnerstag, 03. Juli 2014 um 20:49 Uhr
> Von: WanMil <wmgcnfg at web.de>
> An: "Development list for mkgmap" <mkgmap-dev at lists.mkgmap.org.uk>
> Betreff: Re: [mkgmap-dev] Curvy routing support: new function?
>
> I forgot to clean before build...
> Here is the new version: http://files.mkgmap.org.uk/detail/216
> 
> > Hi WanMil,
> >
> > you are very fast, thank you very much for this!
> > However I'm facing some expectations, log:
> > java.lang.NoSuchMethodError: uk.me.parabola.mkgmap.reader.osm.Way.deleteTag(Ljav
> > a/lang/String;)V
> >          at uk.me.parabola.mkgmap.reader.osm.MultiPolygonFinishHook.end(MultiPoly
> > gonFinishHook.java:50)
> >          at uk.me.parabola.mkgmap.reader.osm.OsmReadingHooksChain.end(OsmReadingH
> > ooksChain.java:79)
> >          at uk.me.parabola.mkgmap.reader.osm.o5m.O5mBinMapDataSource.load(O5mBinM
> > apDataSource.java:49)
> >          at uk.me.parabola.mkgmap.reader.osm.OsmMapDataSource.load(OsmMapDataSour
> > ce.java:127)
> >          at uk.me.parabola.mkgmap.main.MapMaker.loadFromFile(MapMaker.java:167)
> >          at uk.me.parabola.mkgmap.main.MapMaker.makeMap(MapMaker.java:63)
> >          at uk.me.parabola.mkgmap.main.Main$1.call(Main.java:253)
> >          at uk.me.parabola.mkgmap.main.Main$1.call(Main.java:249)
> >          at java.util.concurrent.FutureTask$Sync.innerRun(Unknown Source)
> >          at java.util.concurrent.FutureTask.run(Unknown Source)
> >          at java.util.concurrent.ThreadPoolExecutor.runWorker(Unknown Source)
> >          at java.util.concurrent.ThreadPoolExecutor$Worker.run(Unknown Source)
> >          at java.lang.Thread.run(Unknown Source)
> >
> > (I have installed r3310 which works ok, and then replaced mkgmap.jar by your build)
> >
> > I'm open to any suggestions for other logic. A curviness function can only be a virtual value, not physical. However it may probably sufficient to say, curviness=length()/distance(first_point, last_point)? May be I'm thinking too complex :-)
> >
> > Cheers
> > Manfred
> >
> >
> >
> >> Gesendet: Dienstag, 01. Juli 2014 um 22:32 Uhr
> >> Von: WanMil <wmgcnfg at web.de>
> >> An: "Development list for mkgmap" <mkgmap-dev at lists.mkgmap.org.uk>
> >> Betreff: Re: [mkgmap-dev] Curvy routing support: new function?
> >>
> >> Hi,
> >>
> >> I remember that it is on my Todo-List :-)
> >>
> >> So I performed a quick implementation following your suggestion. You can
> >> find in the attached patch so that you can play a little bit with it.
> >> (Attention: it is completely untested!!)
> >> You can find a patched mkgmap.jar at http://files.mkgmap.org.uk/detail/215.
> >>
> >> I think curviness should to be defined in a different way.
> >> 1. Calculate the distance of each point Dmiddle to the virtual line from
> >> first to the last point.
> >> 2. curviness()=standard mean(Dmiddle/(length to next point + length to
> >> previous point)
> >>
> >> I think this gives a better indication how curviness a road is. But it
> >> need to be implemented and tested later on :-)
> >>
> >> WanMil
> >>
> >>> Hi all,
> >>>
> >>> Garmin offers "curvy roads" preferences for their zümo 390 and 590 devices.
> >>> https://buy.garmin.com/en-US/US/on-the-road/motorcycles/zumo-390lm/prod138275.html
> >>> I'm thinking about creating motorcycle maps for old 200 series.
> >>> Might it be useful to integrate a curviness() function so that mkgmap can optimize for curvy roads too? It might also help do determine a better default speed for curvy roads for use with car routing.
> >>>
> >>> The curviness() value might be the overall absolute turning angle (in degrees) of a road segment divided by the segment's length.
> >>>
> >>> What's your opinion?
> >>>
> >>> Cheers
> >>> Manfred
> >>>
> >>>
> >>>
> >>>
> >>> Pseudo code might look like this:
> >>>
> >>> function curviness()
> >>> {
> >>> oldpoint=points(0)
> >>> foreach (point of a way segment as newpoint)
> >>>     {
> >>>     d_x=(newpoint.x-oldpoint.x)/360*40000000*COS(newpoint.x*PI()/180) //in meters
> >>>     d_y=(newpoint.y-oldpoint.y)/360*40000000 //in meters
> >>>     length+=sqrt(d_x**2+d_y**2)
> >>>     angle=arctan(d_y/d_x)
> >>>     d_angle=abs(angle-old_angle)
> >>>     sgn_x=sgn(d_x)
> >>>     sgn_y=sgn(d_y)
> >>>     if ((sgn_x*sgn_y*old_sgn_x*old_sgn_y)=-1 && (sgn_x!=old_sgn_x))  d_angle=pi-d_angle  // 180 degrees correction, otherwise north-south roads show strange results
> >>>     if (point>1) sum_angle+=d_angle
> >>>     old_angle=angle
> >>>     oldpoint=newpoint
> >>>     old_sgn_x=sgn_x
> >>>     old_sgn_y=sgn_y
> >>>     }
> >>> return sum_angle/length
> >>> }
> >>>
> >>> // perhaps simple multiplication is faster than 4 times sgn() functions?
> >>> _______________________________________________
> >>> mkgmap-dev mailing list
> >>> mkgmap-dev at lists.mkgmap.org.uk
> >>> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
> >>>
> >>
> >> _______________________________________________
> >> mkgmap-dev mailing list
> >> mkgmap-dev at lists.mkgmap.org.uk
> >> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
> > _______________________________________________
> > mkgmap-dev mailing list
> > mkgmap-dev at lists.mkgmap.org.uk
> > http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
> >
> 
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test_curviness.pdf
Type: application/pdf
Size: 88200 bytes
Desc: not available
URL: <http://www.mkgmap.org.uk/pipermail/mkgmap-dev/attachments/20140703/4a689b3b/attachment-0001.pdf>