Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
[sumo-user] How to configure a simulation to obtain the same results in two different runs ?

Hi,
I am doing a simulation using sumo and TraCI, trying to run a benchmark among different traffic lights controllers. For this, I need to reproduce the exact same conditions for car arrival in the network. But I discovered that even running the same simulation twice, with the exact same parameters and input files, I obtain different results. Apparently this happens because sumo includes a lot of random parameters which affect the overall simulation result. I tried to eliminate this randomicity by using --random-depart-offset 0 but, even though the departures were more consistent, there were still some departures with a different departDelay (not so many, it is true, but even there). What should be the keys to append to the simulation to obtain the exact same behavior among two different runs of the same situation ?
Ricardo

Back to the top