All files are read to see if they contain any vehicles that must be inserted. This implies reading at least one vehicle from each file to determine it's departure time. Once this vehicle is read, it draws a random speedFactor from a distribution and thereby influences all other vehicles that are read later. The only way to avoid this is by disabling the speedFactor distribution (i.e. by setting sumo option --default.speeddev 0). However, this makes traffic dynamics unrealistic since all vehicles have the exact same desired speed. There are ways to overcome this but it may be easier to organize the route files differently on your end.
Am So., 5. Juni 2022 um 15:38 Uhr schrieb Kaveh Khoshkhah < [email protected]>: > Hi, > > I ran two simulations and expected the result to be the same. > > 1) sumo -n net.xml -r routes_0.xml --random false --end 3600 > 2) sumo -n net.xml -r routes_0.xml,routes_1.xml --random false --end 3600 > > But I got different results. Depart time of all vehicles in routes_0.xml > is less than 3600, and Depart time of all vehicles in routes_1.xml is more > than 3600. But the simulation results are different. > Is there any solution for avoiding the randomness in simulations? > > I thought maybe the randomness is from TLS. So I added "--tls.all-off > true " to the simulation command. But the results are different again. > The routes format for all vehicles look like this: > <vehicle id="0_10" depart="100" departLane="free" departSpeed="max"> > <route edges="edge1 edge2 edge3 edge4"/> > </vehicle> > > The sumo version is 1.13.0, which I Installed by pip. > > Best regards, > Kaveh > _______________________________________________ > 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
