In version 1.1.0 there is a fix for deadlock caused by long vehicles being unable to re-enter traffic after parking in a short parkingArea. Issue #4832 <https://github.com/eclipse/sumo/issues/4832> (basically, vehicles that wish to enter to parkingArea but are unable to do so because it is full must keep enough space on the road to let others exit) I suspect that the vehicle behavior you are seeing may be related to that change. It would be best if you could send me a small scenario with the original parking area positioning and a reduced set of vehicles to reproduce the problems. You may find the cutRoutes tool useful when transforming large-scenario demand to match a reduced network ( http://sumo.dlr.de/wiki/Tools/Routes#cutRoutes.py) but you will still have to cut the parkingArea definitions with Netedit since cutRoutes.py does not yet support them.
regards, Jakob Am Do., 10. Jan. 2019 um 15:12 Uhr schrieb Lara CODECA < lara.cod...@eurecom.fr>: > Hello Jakob, > > > > Thanks for the suggestions: > > - The issue seems independent from the random seed (I tried with > 15 different seeds and happens all the time) > > - Changing the startPos from 47.09 to 45.09 the problem > disappears. > > > > For the moment I’ll fix it by making the areas larger, but given that my > scenario is “automagically” generated, I would like to discover what’s the > problem, in order to regenerate it right. > > > > In case this tell something to you that it does not for me, the vehicles > have a defined arrivalPos="49.59" but they were stopping exactly at 47.02 > and waiting until teleportation. Given that 47.02 < 47.09 (the beginning of > the parking area) nothing was working. What can it be that influences the > stopping position and changed it from the XML defined one? > > > > All the Best, > > Lara > > > > *From:* sumo-user-boun...@eclipse.org [mailto: > sumo-user-boun...@eclipse.org] *On Behalf Of *Jakob Erdmann > *Sent:* Thursday 10 January 2019 08:16 > *To:* Sumo project User discussions > *Subject:* Re: [sumo-user] Issues with parking area, vehicle stopped on > the road. > > > > Thank you for reporting. > > The way to debug rare events is to figure out whether they are related to > a local feature or to "randomness". > > Does the error disappear or appear somewhere else when running the > simulation with a different random seed? > > If it stays in the same place, tweak the local geometry (parkingArea > positioning and size) and check if it goes away then. > > The end goal is to boil down the problem to a minimal scenario (which I > can then scrutinize for bugs). > > If the problem cannot be replicated with another random seed or a smaller > scenario then I'll take a look at the big version. > > > > best regards, > > Jakob > > > > Am Mi., 9. Jan. 2019 um 15:25 Uhr schrieb Lara CODECA < > lara.cod...@eurecom.fr>: > > I forgot to mention that the exact same files with SUMO 1.0.1 are working > perfectly fine and the issue does not show up. > > > > Regards, > > Lara > > > > *From:* sumo-user-boun...@eclipse.org [mailto: > sumo-user-boun...@eclipse.org] *On Behalf Of *Lara CODECA > *Sent:* Wednesday 09 January 2019 15:13 > *To:* sumo-user@eclipse.org > *Subject:* [sumo-user] Issues with parking area, vehicle stopped on the > road. > > > > Dear All, > > > > I have a problem (that does not happen with all the vehicles) that is: > > > > I have a person > > <person id="pedestrian_2-1_1074" type="pedestrian" depart="18562"> > > <ride from="-152710#1" to="-152902#1" > lines="pedestrian_2-1_1074_tr"/> > > <walk edges="-152902#1 152902#1 152889#2 -153561#0"/> > > </person> > > > > Riding a vehicle > > <vehicle id="pedestrian_2-1_1074_tr" type="ptw" depart="triggered" > departLane="best" arrivalPos="49.59"> > > <route edges="-152710#1 -152710#0 152973#1 152973#2 152973#3 > 152973#4 153441#0 153441#1 153441#2 152468#1 152945#0 152795 152787#2 > 152168 152971#0 152971#1 152971#2 152971#3 -152797#1 152826#0 152826#1 > 152138#7 152138#0 152139 152970 -152271 -152420 -152960 152897#1 152897#2 > 152984#0 152984#1 152984#2 152984#3 -152902#1"/> > > <stop parkingArea="1133" until="43479"/> > > </vehicle> > > > > That has to park. > > > > In sumo 1.1.0 (tarball) this car is stopped in the middle of the road > instead of the parking area. As you can see in the attached image, the > person is still on the vehicle and there is a queue forming behind. > > The stopping point defined for the vehicle is 49.59, and the parking area > begins at 47.09 and ends at 52.09. > > After a very long while (<time-to-teleport value="180"/>), the vehicle > disappears, the parking lot is still empty and the person walks to its > final destination. > > Like this one, all other vehicles that want to park in parkingArea 1133 > get stuck. > > > > All other instances of person+vehicle and parkingArea are defined in the > same way, and they work fine. > > > > All the files are available on GitHub: > https://github.com/lcodeca/MoSTScenario. How can I debug this? Any > suggestions? > > > > Thanks in dvance, > > Lara > > > > --- > > *Lara CODECA* > > Postdoc > > Tel : 04.93.00.81.51 > > > > EURECOM > > Campus SophiaTech > > http://www.eurecom.fr/ > > > > _______________________________________________ > sumo-user mailing list > sumo-user@eclipse.org > To change your delivery options, retrieve your password, or unsubscribe > from this list, visit > https://www.eclipse.org/mailman/listinfo/sumo-user > > _______________________________________________ > sumo-user mailing list > sumo-user@eclipse.org > To change your delivery options, retrieve your password, or unsubscribe > from this list, visit > https://www.eclipse.org/mailman/listinfo/sumo-user >
_______________________________________________ sumo-user mailing list sumo-user@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
