[Tagging] Opinion on meaning of tracktype, smoothness and surface for routing

Fernando Trebien fernando.trebien at gmail.com
Sun Mar 16 21:37:15 UTC 2014


Hello,

Following from this conclusion
(https://lists.openstreetmap.org/pipermail/tagging/2014-March/016904.html),
I'm now trying to find a way to use tracktype, smoothness and surface
to improve routing quality. For an average 4 passenger car (not an
SUV, not a truck, not a motorcycle), I believe that:

1. Maximum "safe speed" is limited by how regular the surface is and
also by how dense the surface material is. The exact material (in the
"surface" tag) is not so important for routing as are these other two
qualities (smoothness and material density).

2. Smoothness and surface density could be somewhat guessed from
"surface" tag in most cases, and smoothness and tracktype could refine
this guess.

The exact "safe speed" for each value is probably a concern of each
application, but for a guess of "expected smoothness" and "expected
tracktype" (in the absence of these tags) in the absence of those
tags, I need opinions. I think these associations make sense when the
other tags are absent:

[format: surface=smoothness,tracktype]
asphalt=excellent,grade1
concrete=excellent,grade1
tartan=excellent,grade1
paved=good,grade1
paving_stones=good,grade1
concrete:plates=good,grade1
metal=good,grade1
compacted=intermediate,grade1
sett=intermediate,grade1
concrete:lanes=intermediate,grade1
bricks=intermediate,grade1
cement=intermediate,grade1
cobblestone=bad,grade1
wood=bad,grade1
stone=bad,grade1
rocky=bad,grade1
gravel=bad,grade2
fine_gravel=bad,grade2
grass_paver=intermediate,grade2
unpaved=bad,grade3
ground=bad,grade3
dirt=bad,grade3
grass=bad,grade3
pebblestone=bad,grade3
clay=bad,grade4
sand=bad,grade5
earth=bad,grade5
mud=very_bad,grade5

What do you think? I know each surface type can vary wildly, but I'm
thinking of those situations in which people probably mean when they
add a surface tag but not a smoothness or a tracktype tag.

I'm not proposing these as defaults for tagging in OSM.

I'm thinking about Emil Tin's idea
(https://www.mail-archive.com/osrm-talk@openstreetmap.org/msg00389.html)
of setting OSRM's speed according to surface as the minimum of several
measures. I tried to multiply two factors as he suggests, but I was
not satisfied with the resulting values. Then I tried to take the
minimum of two speed values associated to smoothness and tracktype and
it seemed much better.

-- 
Fernando Trebien
+55 (51) 9962-5409

"The speed of computer chips doubles every 18 months." (Moore's law)
"The speed of software halves every 18 months." (Gates' law)



More information about the Tagging mailing list