All uploaded solutions are infeasible

Hi,

When I upload my solution xml files, I get errors for all of them. The error messages are of the format:

Planning Problem Solution trajectory does not start at the initial_state of the planning problem with ID: 1 Expected position=[ 399.004 -547.5064] Received position=[ 399.7587933 -548.71238964]

(with differing positional values)

When I test my implementation in the “tutorial_commonroad_search.ipynb” notebook, everything works fine and even the validity-checker returns true for everything. What could cause this error? How am I not starting at the initial_state? Why does the validity-checker return true but the website returns errors?

Thanks in advance

Hi nkRoads,

sorry for the delayed reply, we must’ve missed your post: Please check the answer in this post and see if it solves your issue (Initial state is wrong - #6 by gerald_w). It seems that the problem you stated is similar. If you are still facing problems please let us know.

Best,
Gerald

Hi,

as described in this answer (https://commonroad.in.tum.de/forum/t/initial-state-is-wrong/1085/7) I have the same problem.

After updating the requirements.txt, the initial-state error is now thrown by the local validation checker (before, it was valid locally but invalid on the website). I am running on Mac M1 chip, but this seems to be the same problem as in the other thread.

Thank you for your help.

Hi nkRoads,

I’m not sure why that happens, maybe there is an issue with your local setup: Have you made sure that you are using the latest version of the commonroad-search repository? If not you could pull the latest version of the code. If, in addition the versions are correct according to the requirements.txt then it should produce the correct results locally and on the server.

If you are still having issues, don’t hesitate to reach out.

Best,
Gerald

Hi nkRoads,

it seems that for another user the same issue was solved by making sure the setup is correct, i.e., re-installing the requirements and making sure that the latest version of the code is pulled (Initial state is wrong - #9 by Noah).

Let us know if this also fixes the issue for you.

Best,
Gerald

Hi,

yes it finally works, thank you again for the help.