InternRobotics / InternRobotics/EventVLA
[RoboTwin-MeM] find_seal_and_seal_stamp/demo_clean episode 13 appears to be an erroneous trajectory
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 79
- Forks
- 5
- PR merge metrics
- No merged PRs in 30d
Description
We replayed all 50 released HDF5 trajectories in:
RoboTwin-MeM/hdf5/find_seal_and_seal_stamp/demo_clean
using the released RoboTwin-MeM environment and cached trajectories.
Episode 13 is the only episode where the strict check_success() returns False.
For episode 13:
At around 12 s, the robot arm accidentally collides with a cover. This causes cover 3 to be knocked away later in the trajectory, making the episode fail the task success check.
The other 49 released episodes do not show this issue.
Could you please confirm whether episode 13 should be considered a valid training demonstration?
Thanks for releasing RoboTwin-MeM.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by replaying episode 13 from RoboTwin-MeM/hdf5/find_seal_and_seal_stamp/demo_clean in the released RoboTwin-MeM environment and run the strict check_success() check. Compare its result and trajectory around 12 seconds with the other 49 episodes; done means documenting whether episode 13 is valid or should be excluded, pending maintainer confirmation.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- robotics
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Active
- Clarity
- Needs clarification
- Newbie friendliness
- 38/100