signals using uber turn groups. #555
This was an old half-baked experiment for handling a cluster of traffic
signals. Since then, merging intersections (by manually tagging them in
OSM for now, maybe automatically in the future) has proven better.
Removing this old code in preparation for pathfinding v2.
always tiny; Dijkstra's is fine. It costs a bit of file size to store
it. The huge leeds map goes from 160MB to 157MB -- not crazy savings,
but something.
Also fix a slight bug with 92d3a890ea that
caused some pedestrians to uselessly visit a bus stop node while
routing. (southbank crashes a few hours in otherwise)
This is simpler to reason about, allows the penalty for entering a zone
or taking an unprotected turn to be expressed in terms of a time
penalty, and is a step towards adjusting bike/foot routing for elevation
data.
When we later add things like "safety/quietness" for cycling, maybe we
can switch to using a (time, quietness) tuple, and transform into a
single number with a linear combination parameterized by that agent's
preference for time/safety. This change is compatible with that future
idea.
There are behavior changes here, particularly for zones and unprotected
turns. No new maps start gridlocking, and in fact, Rainier starts
working again.
"rise / run" calculation used the trimmed road center-lines, which don't
match up with the elevation at each original intersection point.
Also handle infinity in the output and reduce the resolution of the
query from every 1m to every 5m.
Regenerate all maps due to the map format change. Try bringing in
elevation data for all of Seattle using the LIDAR source, since
the data quality assessed in eldang/elevation_lookups#12 seems to be
similar, and LIDAR is way faster than contours.
Regenerate all maps. Gridlock-wise, Rainier and Poundbury broke, but
Wallingford started working again. Acceptable cost for a change this
useful; I'll work on fixing those maps later.
Instead of just picking the intersectin closest to the origin or
destination, calculate the full path length, and take the one with the
shortest distance. This fixes some of the weird problems routing around
Broadmoor. Regenerate all prebaked data.
Also fix the original request for paths involving zones, so tracing it
later works.
wind up looping back on themselves in a nonsensical way, causing
vehicles to visually glitch when moving through.
This was started in 081819d86b, but it
used to gridlock 2 maps. All the recent roundabout fixes seems to have
resolved those! And adjusting offstreet parking for two maps.
But wallingford does regress; plunging forward for now.
intersections. They wind up looping back on themselves in a nonsensical
way, causing vehicles to visually glitch when moving through.
This causes lakeslice and rainier to gridlock, due to the magic of
emergent behavior. I think I upstreamed an OSM fix for lakeslice, but I
need to work on rainier before enabling this code.
attention to which intersection is being destroyed. Fixes#527 --
montlake and phinney both look correct now.
Regenerating everything. Actually, Phinney now runs, so adding a 4th
prebaked map!!! But Rainier regressed -- there's an issue with the
signal heuristics that's now a problem; I'll fix later.
I experimented on the Rainier Valley map, which recently started
gridlocking due to too many cars doing this, to tune the value. Got it
running again! The two other maps keep running, with some trips on
average getting a little slower.
Added an extra step to classify service roads as running through a
parking lot, to prevent them from being treated as regular roads.
Had to fix up a few prebaked traffic signals. lakeslice falls back into
gridlock; will fix separately -- too much effort behind this change to
stop.
be more careful with nodes representing uber-turns. Even if that vehicle
type doesn't use an uber-turn, we still need to force the nodes to exist
and match up between input graphs.
Although this really only fixes gb/charleville_mezieres/secteur4, it
potentially affects all maps, because the node map changes. So
regenerate everything...
City names are now disambiguated by a two-letter country code. This
commit handles almost everything needed to make this transition. Main
next steps are fixing up map edits automatically and making the city
picker UI understand the extra level of hierarchy.
A little bit of fallout: lakeslice gridlocks again; this regression is
actually from the recent traffic signal changes, but I'm just now
regenerating everything. Will fix soon.
- Stop alerting when slow pedestrians can't make it through the minimum
crosswalk time
- Simpler iteration style in lagging_green.rs
- Totally delete the old brute force signal config code; it never worked
well, and the improved heuristics eliminate the need for it anyway
- Make a Duration::max function and use it in one case
* Create module from make/traffic_signals.rs
Most things are in mod.rs, lagging_green and brute_force are moved into their own .rs files
* pull more into brute_force.rs
Add a template for lagging green
Lagging green is variable. Crosswalks are also variable.
convert_to_ped_scramble() is refactored to allow a call that doesn't promote yield to protected; as this could unintentionally extend a variable phase.
This lets you live-tune routing parameters, then see how it affects the
total number of routes that cross every road, relative to the baseline
of the original parameters. We can also use it in the future to do the
same thing, comparing before/after some map edits.
* Skip all-walk if no demand
If you create an all-walk stage and make it adaptive, if there is no demand it will be skipped. Fixed crosswalks are never skipped, as they may be necessary for queue management.
u-turns, but recently, they're forced to left turns if the road names
don't match. Now be more specific and revert either a left or right
turn.
This fixes up https://www.openstreetmap.org/node/53089011, which now
acts like a 4-way intersection for signal heuristics!
Not regenerating everything yet.
to match.
Originally these were introduced to deal with merging intersections
between dual carriageways. But inadvertently, lots of left turns got
reclassified as u-turns. That's caused various headaches, most recently
the lakeslice gridlock. That's fixed again!
Fix a bug with the previous commit (lanes=1 on a two-way). Now regenerate.
... Unfortunately lakeslice now gridlocks due to a turn generation bug.
Temporarily removing the prebaked results there so I can push these last
few changes through. Will resolve this before the next release.
Originally, the intention of the deleted calls was to not interrupt
Timer progress bars with warnings. But the output of things like the
importer is impossible to read anyway. Strongly considering explicitly
sending logs and timing info to separate places and using something like
multitail for live progress.
Unplumb timer from LOADS of places that just needed it for logging.
Also give living_streets in Krakow shoulders, so foot routing works
better there.
Now regenerate everything. Actually messes up routing for Trumpington;
71 cancelled trips up to 101. And have to intervene to keep lakeslice
not gridlocking, as usual.
For the moment, this is the simplest way to allow foot traffic. This
breaks down in places like
https://www.openstreetmap.org/way/49207928, where the road gets an
inferred sidewalk and the separate cycleway on each side is
bidirectional with shoulders on each side.
Down to 71 cancelled trips in the baseline for cyipt/actdev#32.
threshold is hardcoded to 0 to disable this, but it can be more easily
changed when experimenting locally.
I tried this out in a few maps manually and saw that it helps sometimes,
but breaks strangely other times. My plan is to work through the bugs
that result and eventually enable this for all imports. #114
geometry), don't generate crosswalks or stop signs. In reality, these
usually represent the middle of a complicatd intersection. Ideally these
cases would be merged into a single intersection, but before that's
feasible, at least improve some of the inferred things nearby. #457
Small adjustments to unzoomed rendering and stop sign placement.
Regenerate all maps because of the format change, but only Cambridge
changes. Since we're doing this anyway, also pull in leisure=garden.
a hard error when they become out-of-date going forward.
Better heuristics make some of these unnecessary. And now the the JSON
files are in this repo, updating files manually when pulling down new
OSM data becomes less tedious.
widgetry, geom, and abstutil may wind up on crates.io in some form to
let other projects use widgetry. abstio has A/B Street-specific tricks
for reading data on native/web. Note widgetry still depends on abstio,
will figure out how to clean that up next.
Originally it was split out to organize a separate volunteer mapping
effort, but that never took off, and it's unlikely to happen. When we
have to occasionally update the prebaked signal data for some
intersections, it's unnecessary friction to update the other repo.
later overriding a .osm file with tags for these roads. This is a way to
test the effects of making simple OSM changes locally before
upstreaming.
Also handle repeated merging collapsing one of the roads. Not 100% what
happens here, but skipping the collapsed road works fine. The scary
Montlake intersections now look great with all the merging done, so
upstreaming the change! #114
everything along Aurora looks fine, but maybe I wrote the other way when
testing in Montlake earlier. Guess I'll find out soon. #114
Not regenerating all maps yet, since more churn is on the way.
Previously, dual carriageways (pairs of one-way roads in opposite
directions) mostly didn't get any signal templates successfully applied.
This change ignores outbound-only roads when applying the templates. In
one fell swoop, lots of previously broken signals along places like
Aurora Ave suddenly work reasonably.
generated incorrectly, but regardless, calling them TurnType::Left is
just confusing.
For the moment, always filter out U-turns from merged intersections.
When connections across merged one-ways are handled properly, we won't
need this, but in the meantime, it moves forward. #114
Not regenerating just yet, but will bundle it with the next commit.
in the interior of a big intersection. #255, #114
- No sidewalks or parking on it
- Automatically try to merge it
Bring in fresh Seattle OSM with a few places on Aurora tagged, for
further experimentation.
Also, there's some bug in the importer; Seattle maps didn't get
regenerated last change. Picking up the diffs now.
For the (still disabled) cases of merging short roads, this helps
immensely. It doesn't affect most other maps visibly. Makes a few
already broken things in Krakow and London slightly worse, but don't
care, because they didn't look sane before either.
map_editor. It was nuked in 182f5139a5,
when I decided the MapFixes approach wasn't worth it. #114
The restored version doesn't attempt to handle turn restrictions again
yet.
* Add a Variable phase
Variable provides a min duration, a delay duration, and an additional duration. The maximum cycle time is min + additional. Once min has been exhausted, if there is demand, the cycle is extended by delay until there isn't any demand or the additional duration has been consumed.
#295
degrees to 30 degrees. It works around the issue in #428, but it doesn't
solve the root cause there, so the unit test is also adjusted to provide
a way to solve the harder problem.
Regenerated all maps accordingly. Many traffic signals tended to
improve, with a straight turn marked protected, instead of permitted as
a "right turn."
saves lots of callers from cloning the request and separately plumbing
around the requested start/end distance. Also a step towards exposing
more granular distance crossed in a path for #392.
Still a few more places to simplify, but will do in a separate, smaller
change.
1. Allow lane changing in an uber turn. Because of the way uber turns
work, we lock in and commit to all the lane changes just before
entering the uber turn.
2. Avoid overzealous lane changing by combining number-of-lanes-crossed
and numer-of-vehicles-in-lane into a single cost, rather than always
preferring the least number-of-vehicles-in-lane.
3. Don't lane-change unless the candidate lane's cost is strictly better
than the current lane cost.
It'd be more efficient to terminate the Dijkstra's search directly, but
petgraph doesn't have an option for that, so we'll have to implement
Dijkstra's manually (shouldn't be hard).
struct. Whatever choices we make next about naming cities hierarchially
or not can be managed in just one place. #326
This is a pretty huge change, but the compiler gives reasonable
confidence it's correct. More bugs are likely to crop up in the next
step, when filenames start being namespaced by the city too.
