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

Use handover detector to detect failed handover #1327

Merged
merged 6 commits into from
Aug 8, 2023

Conversation

PetervDooren
Copy link
Contributor

Merge after #1326
Fixes one of the issues in #1325

@PetervDooren PetervDooren mentioned this pull request Jul 25, 2023
3 tasks
@PetervDooren PetervDooren requested review from shanki98 and removed request for MatthijsBurgh July 25, 2023 17:40
Copy link
Contributor Author

@PetervDooren PetervDooren left a comment

Choose a reason for hiding this comment

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

Tested on the robot. Worked like a charm

Copy link
Member

@MatthijsBurgh MatthijsBurgh left a comment

Choose a reason for hiding this comment

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

I think the issue was always it requires tuning per item it is holding. I think that downside is still there. So works like a charm for a subset of items probably. Please prove me wrong.

@PetervDooren
Copy link
Contributor Author

That subset is all objects that are not paper thin. Which includes at least all objects used in serve breakfast. I tested it with a spoon and bowl

Copy link
Member

@MatthijsBurgh MatthijsBurgh left a comment

Choose a reason for hiding this comment

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

Add comment about the used tuning for detector works for what set of items, i.e. their dimensions

@PetervDooren
Copy link
Contributor Author

There was no tuning done in this PR. it just uses the grasp detector. Grasp detector is tuned for how accurate it can detect things. No need to add a comment here.

@MatthijsBurgh
Copy link
Member

There is no comment either in the challenge or the ActiveGraspDetector about which the dimensions is works on. It is only mentioned when the object is too thin/small it doesn't work. No clue about what that threshold is.\

@PetervDooren
Copy link
Contributor Author

Docstring of ActiveGraspDetector states anything wider than the threshold (0.075 rad on the gripper) can be detected. You can say this is not clear enough but that is not an issue for this PR

   State for detecting whether the robot is holding something using the gripper position.

    Stores current position of the hand motor joint, slightly closes the gripper and then compares the new
    position with the first one. If the difference is bigger than the threshold, robot is holding something

    If the object is too small/thin it will not be able to determine

    :param robot: Robot to execute the state with
    :param arm_designator: designator that resolves to arm to check
    :param threshold_difference: Difference between base and final position
    :param minimum_position: Minimum position to assume that the gripper is holding something
    :param max_torque: Max torque of the gripper to perform the test with

MatthijsBurgh
MatthijsBurgh previously approved these changes Aug 8, 2023
@PetervDooren PetervDooren merged commit 092a354 into master Aug 8, 2023
@PetervDooren PetervDooren deleted the serve_breakfast/handoverdetector branch August 8, 2023 18:11
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants