Most likely, the first vehicle did not clear up the insertion location quickly enough for the second vehicle to enter on time *safely*.
Where *safely* is related to the gap that the car following model would like to maintain at insertion based on the desired insertion speed.
Hi, Jakob,
I simulated several continuous intersections on sumo, and the depart time of the vehicles subjected to uniform distribution (the depart time equal to 4s in my experiments). I used reinforcement learning algorithm to generate the acceleration for the vehicle at each simulation step. What confuses me is that the output of the routing file shows that the depart delay is not equal to 0 for some vehicles. (The desired depart time of the second veh should be 8 s, but the actual depart time is 9 s, depart delay = 1 s). I would like to know how to solve this problem.
As always, many thanks !
Houjia
从 Windows 版邮件发送
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit
https://www.eclipse.org/mailman/listinfo/sumo-user