Hi,
I am currently working on solving interactive scenarios and have generated several solutions. When I try to save the solution, it comes:
“assert pp_id in planning_problem_set.planning_probelm_dict”
I believe this is trying to check whether the planning problem of the ego vehicle matches that of the scenario by comparing pp_id. However, I find their pp_id (namely, dict.key()) do not match right after I generate ego_vehicles from the scenario_wrapper by running SumoSimulation(). See as follows:
(from: DUE_Frankfurt-4_2_I-1)
What should I do to fix it?
By the way, Is there any detailed description of the solution checking criterion? We manually check invalid solutions (labeled by valid_solution()) by watching the videos at present, but sometimes videos look pretty good which makes it difficult to debug.
Thanks a lot for your help!
Desmond
Hello Desmond,
when running the interactive scenarios with this function, the returned dict ego_vehicles has the pp_id as the key, which is also used when saving the solution files with the function save_solution. You can find this function also in the tutorial. Probably you took the id from the attribute EgoVehicle.id? This id is indeed different from the pp_id.
To debug the reason for the failure you might use the individual functions called by valid_solution() and in particular trajectory_feasiblity. Maybe this tutorial is also helpful for you since it already provides visualizations: https://commonroad.in.tum.de/tutorials/04_feasibility_checker. You can download it from the repository. The point at which the check fails can be obtained if you set a breakpoint here. As mentioned in this comment, the motion might be too abrupt.
That helps a lot!
Thanks for your timely reply.