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

How to print the tags. #12

Closed
Mikor-mkr opened this issue Sep 22, 2021 · 3 comments
Closed

How to print the tags. #12

Mikor-mkr opened this issue Sep 22, 2021 · 3 comments

Comments

@Mikor-mkr
Copy link

Hello,

thanks for taking the time to check out my post. I want some help for printing the targets, more specifically is unclear to me how to generate those tags. Please, can you provide some information on that? I see that you provide some information about the targets at the lib folder but it is unclear tome how to use it. Also, I understand that I need to have matlab installed in order to print custom tags, is that right?

Thanks in advance!

@brucejk
Copy link
Member

brucejk commented Sep 23, 2021

Hi,

Thank you for being interested in this package! I will try my best to help. I would like to clarify the question.

  1. When you said "generate those tags", do you mean you want to use a different type of fiducial marker (not using ApriltagTag) for LiDARTag?
  2. When you said "print custom tags", do you mean you need to regenerate the function library for the new fiducial markers for the LiDARTag?
  3. Specifically, are you asking about Building Your Own Customized Markers?

@brucejk
Copy link
Member

brucejk commented Oct 6, 2021

I am going to close this issue due to inactivity. Please feel free to reopen if you have similar questions.

@brucejk brucejk closed this as completed Oct 6, 2021
@mmalmeid
Copy link

mmalmeid commented Feb 8, 2023

Since I have been through the same issue, here is how I generated pdf's of appropriate size fiducials in the family 16h06.
Just download the resize-tags.sh script from this link, and place it in LiDARTag/matlab/function_dictionary. Then update the configs of the script following the instructions in the link above, and you should have pdf's and png's with the for the 16h06 with the appropriate sizes

brucejk pushed a commit that referenced this issue Mar 3, 2023
* Changes to fix the remappings.

Also deleted unused or redundant code/parameters

* deleted unused members in structures and unused code

* The lidartag was using pandar as the lidar_frame in a hardcoded fashion.
Since we plan to use several pandar instances with different names, we must parameterize this frame from the calibration tools

* Fixed several errors and implemented several improvements to increase the detection rate and precision of the detections.
 - Fixes clusters that could not be formed correctly in one pass
 - Adds an optional rectangle estimation rather tha 4x line estimation to get the corners, it is much more stable and increases both the accuracy and detection rate
 - Refines the initial center of the tag based on the corner fitting which seems to improve the pose estimation. In many cases the nlopt optimizator is not even needed

* - Calibration tools compliance
 - Fix edge groups provided by PCA using the partial rectangle models produced by the RANSAC iterations. Provides great results.
 - Some debug drawings to help understand the previous fix
 - Parameterized min decoding score

* Truned off the verbosity that is lidartag targetting its release

* Updated linkage

* Added empty file because git does not support empty folders

* Added logic to detet lidartags without intensity ad id=-1
Fixed logic error that cuases entire rings to become invalid when the tag was high enough to confuse the system (this was an old bug that escaped detection several times).
Added launcher logic to use both pandar_40p and pandar_qt.
Calibrated the system for pandar_qt use, but there are errors in the measures of the lidars

* Fixed the detection pose in the msg

* Added the launcher for the xx1 velodyne

* Added a profile for the velodyne vls128 lidar
Parameterized the maximum ring difference criteria to add points to a cluster (usually should not be needed but one of our lidars was missing a channel so this acts as a hotfix)
Fixed the reflection problem due to the naive use of the SVD method to estimate rotations

* Implemnted review changes

* Removed comments

* Removed the use of internal eigen.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Development

No branches or pull requests

3 participants