Hello, that does look strange indeed. Unfortunately, I have no idea what could be going on there. Can send me the files for the "classic control" scenario? Have you looked at the microscopic behavior and found any clues (i.e. unmotivated large gaps between vehicles)?
2015-08-04 16:21 GMT+02:00 <[email protected]>: > Dear Jakob, > > we are using a static route file. TraCI is not used for vehicles insertion > in our case. > I don't know if traffic dynamics also plays a role but i suppose no. > > I have plotted some graphs that compare sumo 0.19 and sumo 0.23 in two > kinds of control within the same static route file and network. > It seems inserted vehicles differ in sumo 0.23 between the two controls > (see figure 5). > > What do you think about this? > > Best regards > > Cyril > > ------------------------------ > > *De: *"Jakob Erdmann" <[email protected]> > *À: *"Cyril Nguyen Van Phu" <[email protected]> > *Cc: *"sumo-user" <[email protected]> > *Envoyé: *Mercredi 29 Juillet 2015 15:58:32 > *Objet: *Re: [sumo-user] vehicles insertion in sumo > > We had some insertion related bug-fixes, some of which should increase > maximum possible flow whereas others might conceivably lower it. It's hard > to tell what's going on in your simulation without further details. > - are you using a static route file or adding the vehicles dynamically via > TraCI? > - are insertions the root cause of the difference between 0.21.0 and > 0.19.0 or could different traffic dynamics also play a role? > - how do the scenarios behave using sumo 0.23.0? > > If you can demonstrate a problem in the latest release (0.23.0) such as > vehicles not being inserted even though there is sufficient space, we will > try to fix it. > > regards, > Jakob > > > > 2015-07-27 12:33 GMT+02:00 <[email protected]>: > >> Hello sumo users, >> >> while running simulations with sumo-0.21 via traci, I've noticed that >> vehicles insertion differ from sumo 0.19. >> >> Indeed, we developed a new kind of traffic light signal control and >> compared it to another existing control. >> >> In sumo 0.19 we noticed that there is a congestion in both controls with >> a scenario where the demand is high. >> But in sumo 0.21, Within the same scenario, we can't reach any congestion >> in one control although there is some congestion in the other control. >> >> By examining the simulation output trace file, we noticed that less >> vehicles are inserted within one control in sumo 0.21 which explains the >> fact that there is no congestion in this case. >> >> Is this problem known? >> Is there any explanation about this problem? >> >> Best regards, >> >> Cyril >> >> ------------------------------------------------------------------------------ >> _______________________________________________ >> sumo-user mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/sumo-user >> > > ------------------------------------------------------------------------------ _______________________________________________ sumo-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/sumo-user
