Thank you Jakob, it worked wonders ! My optimization is now 25 times faster !!

Yann

De : Jakob Erdmann [mailto:[email protected]]
Envoyé : lundi 6 janvier 2014 14:55
À : SEMET Yann
Cc : [email protected]
Objet : Re: [sumo-user] Average stopping time without TraCI

Hello,
you can compute this metric using edgeData output and summary-output
http://sumo-sim.org/wiki/Simulation/Output/Lane-_or_Edge-based_Traffic_Measures
http://sumo-sim.org/wiki/Simulation/Output/Summary
the edgeData output gives you the aggregated waitingTime for each edge 
(aggregated over all vehicles on the edge during a configurable time period). 
The summary output gives the number of active and arrived vehicles for each 
simulation step.
assuming edgeData is set to aggregate 5 min. intervals (freq="300") then you 
can parse edgeData output for the period from second 0 to second 300 and sum 
waiting time for all edges. Divide this by the sum of active + arrived vehicles 
at second 300 and you have your average stopping time per vehicle.
regards,
Jakob
PS: parsing sumo output files can easily be done using sumo tools:
either use a python script such as:

from sumolib.output import parse_fast
file = sys.argv[1]
print sum(map(float, [e.waitingTime for e in parse_fast(file, 'edge', 
['waitingTime'])]))
or convert the output xml to a csv file and continue from there (see 
http://sumo-sim.org/wiki/Tools/Output#xml2csv.py)




2014/1/6 SEMET Yann 
<[email protected]<mailto:[email protected]>>
Hi all,

I'm using Sumo for a traffic lights optimization project.

I'm currently taking over someone else's code for the Sumo-based simulation 
module. We're using the average stopping time metric as a measurement of 
traffic fluidity and to that end, the other developer was forced to use TraCI 
instead of xml files to exchange data with Sumo (TL programs in, vehicle logs 
out). The reason was that, I think, the information wasn't complete in the 
regular xml output to compute the metric.

The problem is that the optimization loop is way way slower with TraCI than it 
was with simple file based data communication (around 30 times slower). It's 
critical because we want to make as many calls as possible to Sumo  to achieve 
good optimization within a limited time window.

Hence my question(s): can one compute the average stopping time metric (average 
over all vehicles having been or still being present during the simulation for 
a given time window) ? I have two options in mind: a) modifying Sumo options so 
one has everything needed in the output files to compute the metric from 
scratch and b) somehow retrieving this admittedly classical metric already 
calculated by Sumo...

Thanks in advance,

Yann Semet
------------------------------------------------------------------------------
Rapidly troubleshoot problems before they affect your business. Most IT
organizations don't have a clear picture of how application performance
affects their revenue. With AppDynamics, you get 100% visibility into your
Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk
_______________________________________________
sumo-user mailing list
[email protected]<mailto:[email protected]>
https://lists.sourceforge.net/lists/listinfo/sumo-user

------------------------------------------------------------------------------
CenturyLink Cloud: The Leader in Enterprise Cloud Services.
Learn Why More Businesses Are Choosing CenturyLink Cloud For
Critical Workloads, Development Environments & Everything In Between.
Get a Quote or Start a Free Trial Today. 
http://pubads.g.doubleclick.net/gampad/clk?id=119420431&iu=/4140/ostg.clktrk
_______________________________________________
sumo-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/sumo-user

Reply via email to