eclipse-sumo
Vehicles block each other in front of parking vehicles
Hi, I am running a SUMO simulation with a set of parked vehicles on road and it seems like some vehicles are showing a weird behavior when changing lanes around parked vehicle. They also move into the space of parking area which is already occupied by a vehicle which ideally should not happen. I have attached a video below. Appreciate if you can look into this. Thank you.
https://github.com/eclipse-sumo/sumo/assets/55861474/46654045-cb1c-4aa8-857b-7a7ba7ef451b
SUMO-version: 1.18
operating system: Windows 10
I noted:
- some lane change issues due to the high flow (vehicles don't manage to change lane before the parking lane begins)
- vehicle queues grow upstream into empty parking spaces
I didn't see vehicles entering an already occupied parking space... Maybe can you point us to the exact situations you consider weird?
I have attached an image to describe what I mean by vehicles entering an already occupied parking space. I think if the vehicle parks in the middle of the bay the impact of this would be reduced. In addition to that, the bullet points you have listed are the weird behaviors I have observed. Two vehicles get stuck behind the parking area, yet there is no reason for them since they are not looking for parking and they should move ahead using the other lane.
Right, I didn't pay enough attention. Can you attach the zipped scenario to this issue? The vehicles shouldn't get stuck like that.
I have attached the files here. Thank you for looking into this.
Hi, thank you for identifying this as a bug. Would this be assigned to someone from the development team?
@namdre Hi there, I'm conducting research on street parking using SUMO and I have been observing this issue frequently. I am quite aware that you guys have a lot of improvements and bug fixes planned for the platform and you guys are doing a great job. Yet is there any chance we could get this fixed soon? It would be highly appreciated if so and I am more than happy to help you guys with any sort of debugging and fixing that could be done from my end.
Appreciate the offer to contribute. It's easier for me to tackle this issue if you can reproduce it in the development version. Right now, there are no deadlocks/teleports when using the current nightly build: https://sumo.dlr.de/docs/Downloads.php#nightly_snapshots
Oh shoot, I missed this comment. I'll use the nightly build and let you know. Maybe the issue is already solved. Thank you.
@namdre With the nightly build I could not observe the issue occurring. Probably it has been solved by some other developments. Let's close the issue for now and I'll reopen it if I observe it in the future. Thanks a lot for your support on this.
@namdre Unfortunately, I observed the issue again with version 1.20. I have attached a video below. Would you be able to reopen the issue and look into this? Thanks a lot.
https://github.com/eclipse-sumo/sumo/assets/55861474/8ee5982a-da43-44f5-8150-6b74dbd48b28
You can reproduce the issue by running "sumo-gui -c melb.sumocfg --tripinfo-output test.xml --step-length 0.1" with the attached files. Kindly let me know if you need any additional information or support. Thank you.