Fixes action reset of pre_trained_policy_action #1623
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
nicolaloi
requested review from
Dhoeller19,
Mayankm96,
jsmith-bdai,
jtigue-bdai and
kellyguo11
as code owners
hapatel-bdai
pushed a commit
that referenced
this pull request
Jan 21, 2025
# Description Currently, the [PreTrainedPolicyAction](https://github.com/isaac-sim/IsaacLab/blob/v1.4.0/source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/navigation/mdp/pre_trained_policy_action.py#L24) class does not reset the actions in the low-level observations when a new episode starts. In my custom legged robot navigation task, the behavior was correct only during the first training episode but failed from the second episode onward. At the start of a new episode, the action observations are not reset and retain the last actions from the previous episode. This can impact training, as in my case, where the actions at the end of an episode differ significantly from those required at the beginning of an episode. This PR resolves the issue by resetting the low-level action observations at the beginning of each new episode. ## Type of change <!-- As you go through the list, delete the ones that are not applicable. --> - Bug fix (non-breaking change which fixes an issue) ## Checklist - [x] I have run the [`pre-commit` checks](https://pre-commit.com/) with `./isaaclab.sh --format` - [ ] I have made corresponding changes to the documentation - [x] My changes generate no new warnings - [ ] I have added tests that prove my fix is effective or that my feature works - [x] I have updated the changelog and the corresponding version in the extension's `config/extension.toml` file - [x] I have added my name to the `CONTRIBUTORS.md` or my name already exists there <!-- As you go through the checklist above, you can mark something as done by putting an x character in it For example, - [x] I have done this task - [ ] I have not done this task -->
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Currently, the PreTrainedPolicyAction class does not reset the actions in the low-level observations when a new episode starts.
In my custom legged robot navigation task, the behavior was correct only during the first training episode but failed from the second episode onward. At the start of a new episode, the action observations are not reset and retain the last actions from the previous episode. This can impact training, as in my case, where the actions at the end of an episode differ significantly from those required at the beginning of an episode.
This PR resolves the issue by resetting the low-level action observations at the beginning of each new episode.
Type of change
Checklist
pre-commitchecks with./isaaclab.sh --formatconfig/extension.tomlfileCONTRIBUTORS.mdor my name already exists there