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.

Sign up for GitHub

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

Jump to bottom

Update RL examples: Drive and VehicleFollowing #2061

Merged
merged 6 commits into from
Jun 12, 2023
Merged

Update RL examples: Drive and VehicleFollowing #2061

merged 6 commits into from
Jun 12, 2023

Conversation

Adaickalavan
Copy link
Member

Updated RL examples: Drive and VehicleFollowing

  • RGB image fed into the neural network uses higher resolution and is cropped to focus on frontal view.
  • If mission goal is present, the goal position is superimposed onto rgb image before being fed into the neural network.
  • Refactored the RL training script.
  • Training scenarios for Driving SMARTS 2023.1, 2023.2, and 2023.2 benchmarks are updated in the docs, in the benchmarks, and in the training examples.

@Adaickalavan Adaickalavan self-assigned this Jun 10, 2023
This was referenced Jun 11, 2023
Merged
Merged
Gamenot
Gamenot approved these changes Jun 12, 2023
View reviewed changes
Copy link
Collaborator

@Gamenot Gamenot left a comment •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good work as always.

examples/rl/platoon/inference/__init__.py Outdated Show resolved Hide resolved
examples/rl/platoon/inference/contrib_policy/filter_obs.py
@@ -75,6 +91,28 @@ def filter(self, obs: Dict[str, Any]) -> Dict[str, Any]:
if all(rgb_ego[img_y, img_x, :] == self._no_color):
rgb_ego[img_y, img_x, :] = self._wps_color

# Superimpose goal position onto rgb image
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This seems like a useful feature to add to scenario generation or the camera configuration.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Adding the goal position into Envision will make our simulator more visually attractive.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I have repurposed #419 which had a rough intention.

saulfield
saulfield approved these changes Jun 12, 2023
View reviewed changes
examples/rl/platoon/train/reward.py Outdated Show resolved Hide resolved
@Adaickalavan Adaickalavan merged commit 1ad324b into master Jun 12, 2023
@Adaickalavan Adaickalavan deleted the adai/update-rl-example branch June 12, 2023 21:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@Gamenot Gamenot Gamenot approved these changes

@saulfield saulfield saulfield approved these changes

@qianyi-sun qianyi-sun Awaiting requested review from qianyi-sun

Assignees

@Adaickalavan Adaickalavan

Labels
release critical
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@Adaickalavan @Gamenot @saulfield