Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Bug]: Agent doesnt drive after update to Leaderboard 2.0 #76

Closed
1 task
samuelkuehnel opened this issue Nov 20, 2023 · 4 comments
Closed
1 task

[Bug]: Agent doesnt drive after update to Leaderboard 2.0 #76

samuelkuehnel opened this issue Nov 20, 2023 · 4 comments
Assignees
Labels
bug Something isn't working help wanted Extra attention is needed

Comments

@samuelkuehnel
Copy link
Collaborator

Current Behavior

After update to Leaderboard 2.0 in #37 the agent doesnt start to drive.
Apparantly the start routine is not executed.

Expected Behavior

  • Agent drives as already known from the previous version

How to reproduce the issue

  • Get branch 37-bug-update-docker-image-for-carla-simulator-to-fix-build-process
  • Run b5 install and b5 run
@samuelkuehnel samuelkuehnel added the bug Something isn't working label Nov 20, 2023
@samuelkuehnel samuelkuehnel self-assigned this Nov 20, 2023
@samuelkuehnel samuelkuehnel moved this to 📋 Backlog in PAF Project Backlog Nov 20, 2023
@samuelkuehnel samuelkuehnel moved this from 📋 Backlog to 🏗 In progress in PAF Project Backlog Nov 21, 2023
@samuelkuehnel
Copy link
Collaborator Author

Current problem is that the coordinates recieved from our GNSS do not match the coordinates expected from the route:

Image

In the leaderboard 1.0 the coordinates match.

Image

It seems like the calculations done in coordinate_transformation.py are wrong as they depend on the map.

@samuelkuehnel
Copy link
Collaborator Author

The values received from our GNSS seem to be different anyway in the new leaderboard version:

Graph from leaderboard 1:

Image

Graph from leaderboard 2:

Image

@samuelkuehnel
Copy link
Collaborator Author

Update: The reason for this was found in the coordinate_transformation.py file. The calculation there is built on reference coordinates that seem to be the map origin (0, 0, 0). For Town 12 the reference coordinates were set to different values, what produced wrong coordinates in the new leaderboard.

Now the calculated coordinates are calculated correctly and a trajectory is created.

However the car still doesn't follow the trajectory and steers onto the sidewalk where it gets stuck. Next step is to compare the new trajectory to the old one to verify if the issue is located in the calculation of the trajectory or in the acting components

@MaxJa4
Copy link
Collaborator

MaxJa4 commented Nov 23, 2023

Now the calculated coordinates are calculated correctly and a trajectory is created.

Can this be calculated independent from the map or is it hardcoded now?
Since we have Town 12 & 13 to test drive and unknown T14 for validation.

@samuelkuehnel samuelkuehnel moved this from 🏗 In progress to ✅ Done in PAF Project Backlog Jan 30, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working help wanted Extra attention is needed
Projects
Archived in project
Development

No branches or pull requests

3 participants