| Re: [sumo-user] Building a scenario without knowledge of routes, based on flow measurements |
Thanks for your help Jakob! I took your hints into account, but this method is unfortunatly not appropriate for my given data.
I have one more question though. My Network is a small part of a City and I only have traffic data for some of the edges. I still kind of know where the main flows go and where the edges have the function of a sink/source or via or both.
Is there any method that might be more approriate then the one with random.py and calibrators?
I was also triyng to use flowrouter.py, but it always gives me the message "AttributeError: module 'detector' has no attribute 'DetectorReader'" when I want to make it work. I have not read of any Attribute named this way so far, so that I don´t know what to do.
Kind regards,
Florian
Von: sumo-user-bounces@xxxxxxxxxxx <sumo-user-bounces@xxxxxxxxxxx> im Auftrag von Jakob Erdmann <namdre.sumo@xxxxxxxxx>
Gesendet: Mittwoch, 13. Juni 2018 22:52
An: Sumo project User discussions
Betreff: Re: [sumo-user] Building a scenario without knowledge of routes, based on flow measurements1) randomTrips.py can generate vehicles with fixed routes (--route-file) as well as vehicle trips that only specify origin and destination (--output-trip-file). If you use the former, vehicles will pick the fastest route assuming an empty network. If you use the latter, vehicles will use the fastest route considering the state of the network at the time of departure.
2) randomTrips.py accepts the --vehicle-class option to generate vehicles with that specified class. Just run randomTrips multiple times with different options. Make sure to also set the --prefix option to avoid name clashes. I Advise to use osmWebWizard.py to generate a scenarion with different vehicle classes and use the generated files (including batch files that call randomTrips) as a reference: http://sumo.dlr.de/wiki/Tutorials/OSMWebWizard
3) Calibrators do not spread the inserted traffic evenly to allow for vehicles to arrive later during the calibration interval. To avoid this effect you need to define multiple calibrators that cover shorter time intervals
regards,Jakob
2018-06-13 17:15 GMT+02:00 florian preusse <flopreusse@xxxxxxxxxx>:
Dear Sumo users,
i have a few questions to the method "Building a scenario without knowledge of routes, based on flow measurements".
First of all I created random demand and wanted to ask if it is somehow possible to:
1) influence the generated routes/trips, so that the fastest or shortest route would be taken and no Loops would be made within the network. Kind of like the functions that Duarouter offers.
2) generate more than just one vehicle class, so that there would be passenger cars and trucks for example.
When it comes to the calibrators integrated in the network to make sure that only a specific amount of traffic enters a certain edge, I have one question:
1) Is it possible to for the Calibrators to constantly work the whole time of the Simulation?
Right now my calibrators function like this: They dont do anything at the begin of the Simulation and then close to the end of the Simulation they begin to insert or take away vehicles to meet the specific amount, but this is highly unrealstic.
Thank you for your help!
Kind regards,
Florian
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user