[mkgmap-dev] Problem with --adjust-turn-headings
From Mark Burton markb at ordern.com on Tue Feb 9 08:56:56 GMT 2010
Hi Apollinaris, > found a problem in the osm data. there was a short bridge with the name of the exit instead of the motorway name. the turn instruction was exactly as expected according to your implementation. OK. > I have another interesting one here > http://maps.google.com/maps?f=q&source=s_q&hl=en&geocode=&q=sacramento&sll=37.0625,-95.677068&sspn=58.294644,106.699219&ie=UTF8&hq=&hnear=Sacramento,+California&ll=38.573956,-121.578934&spn=0.000448,0.000814&t=k&z=21&layer=c&cbll=38.573604,-121.581104&panoid=e4tOr0puxhW-8sDps67wRA&cbp=12,89.03,,0,18.64 > or in osm > http://www.openstreetmap.org/?lat=38.57493&lon=-121.5694&zoom=16&layers=B000FTFT > Starting at this position I-80 has 3 lanes, to continue on I-80 a keep right instruction is required. But because of same name and same ref no instruction is given. Now adding a through_route relation is an option. But the straight route is also named I-80 BUS and I think in this case an instruction should be given for both routes because the importance is equal for both routes > Is it possible at all to do this in Garmin format? Some Garmin GPS units can tell you which lane to be in (lane assist?) but we don't know how to encode that info. Pity, as that's just what you need here. But looking at the sat pic, I would think that adding a through_route relation is reasonable because more of the road lanes continue on to the I-80 BUS than they do to the I-80. Ah, a subtle point here is that it will probably tell you to turn right rather than keep right at the junction. But I can probably fix that by making mkgmap notice that all of the roads are motorways and so reduce the amount of heading change. But in the meantime, if you want to add that relation (and you will need my patch to mkgmap, of course) and test what happens, it will be interesting to know. > > And again there is a navi failing this test. Google navi on Android doesn't announce a keep right instruction and I bet 50% who rely on their navi will miss the right turn. Very likely, so we should be able to do better than that! Cheers, Mark
- Previous message: [mkgmap-dev] Problem with --adjust-turn-headings
- Next message: [mkgmap-dev] Problem with --adjust-turn-headings
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the mkgmap-dev mailing list