[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
Re: [sumo-user] device.rerouting
|
Thank you Jakob. This piece of information is gold in my model.
Many thanks
Mohsen
On Fri, Sep 28, 2018 at 10:55 PM, Jakob Erdmann
<namdre.sumo@xxxxxxxxx> wrote:
An important aspect of adaptation-steps is to smooth traffic light cycles.
Thus, the product of adaptation-interval and adaptation-steps should be at least as big as the longest traffic light cycle.
If there are cycles with 60s and 90s in your network I would use something like 180 (least common multiple).
Since updating is a bit expensive, a cheap way to get this smoothing effecty would be to use
adaptation-interval 10
adaptation-steps 18
The bigger the value, the bigger the smoothing effect obviously. Wether this is desireable depends on the context.
The period parameter determines how fast vehicles can react to traffic fluctuations. Since rerouting is quite expensive I would not set this too low. We often use 300 in our own scenarios. It should definitely be higher than adaptation-interval because the values don't change in between updates.