Thank you for providing the example files. The crash was caused by a particular road configuration and is now fixed: https://github.com/eclipse/sumo/issues/9718 KraussPS was not directly responsible for the crash but it did highlight the problem that opposite overtaking did not take slopes into account yet (now fixed:https://github.com/eclipse/sumo/issues/9719)
I recommend upgrading to the latest development version which contains the fix starting tomorrow at https://sumo.dlr.de/docs/Downloads.php#sumo_-_latest_development_version regards, Jakob Am Do., 2. Dez. 2021 um 10:40 Uhr schrieb Schaniel Joel < [email protected]>: > Good morning, > > > Thank you for the quick response. > > I tried to reduce the complexity of the simulation as much as possible. > Since it is unclear what is causing the problem, I could not reproduce the > error with a smaller subset of the simulation. > > The KraussPS version crashes at 296 seconds. > > > Best, > > Joel > ------------------------------ > *Von:* sumo-user <[email protected]> im Auftrag von Jakob > Erdmann <[email protected]> > *Gesendet:* Mittwoch, 1. Dezember 2021 20:30:11 > *An:* Sumo project User discussions > *Betreff:* Re: [sumo-user] Sumo crashes because of KraussPS car-following > model > > It's a bug. Please provide a minimal input example for reproducing the > crash. > > Am Mi., 1. Dez. 2021 um 16:56 Uhr schrieb Schaniel Joel < > [email protected]>: > >> Hi everybody, >> >> >> I work on a network in hilly terrain with very steep roads. Therefore, I >> included elevation data in my network file with --heightmap.geotiff. >> Netconvert generates some warnings in the process, which might be related >> to the problem: >> >> >> >> * Warning: 22 total messages of type: Connection '%' has a grade of %%. >> Warning: 2280 total messages of type: Edge '%' has a grade of %%. * >> >> Here is my first question: Why do these warnings appear, even though >> --geometry.max-grade.fix >> is activated? Or are these warnings just generated before the edges get >> fixed and I can simply ignore them? >> >> However, I tried to run the simulation with these warnings. I use the >> "KraussPS" car-following model to include the slope impacts on >> acceleration. But when I run it, the simulation (and also sumo itself) >> crashes always at the same simulation step without any error message. (with >> sumo-gui as well as with sumo) >> >> If I run the same simulation with the normal "Krauss" car-following >> model, the simulation runs through without any problems. >> If I run the same simulation without elevation data, KraussPS works also >> fine (obviously). >> Any idea why this could be? >> I tried to understand how the KraussPS car-following works, but I did not >> manage to apprehend it completely, to be honest. From the comments in the >> code, I understand that there are mechanisms in place that prevent vehicles >> from stalling or drastic changes in the vehicle speed. >> >> I work with a Nightly "extra" Build of version 1.11.0 (17.11.21) on Win10. >> >> Thanks a lot for taking the time! >> >> Best regards, >> Joel >> >> _______________________________________________ >> sumo-user mailing list >> [email protected] >> To unsubscribe from this list, visit >> https://www.eclipse.org/mailman/listinfo/sumo-user >> > _______________________________________________ > sumo-user mailing list > [email protected] > To unsubscribe from this list, visit > https://www.eclipse.org/mailman/listinfo/sumo-user >
_______________________________________________ sumo-user mailing list [email protected] To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
