Closed osrf-migration closed 9 years ago
Original comment by Nate Koenig (Bitbucket: Nathan Koenig).
Do you have more information? We don't have an Atlas, so any little bits will help.
Original comment by Mathew DeDonato (Bitbucket: mdedonato NA).
In v5 simple shapes the head joint has an origin of the following
This causes the model to visibly be off from the model. If you assume the v4 simple shapes has the correct joint setup for the sensor head, you need to update this line to
I am not sure why the offsets were in there.
Original comment by Bill Blank (Bitbucket: BillBlank).
The neck_ry offset file changes were introduced Tue 25, just a few days after after the first AtlasSimInterface3.0.0 was shipped. John Hsu was focusing on resolving discrepancies between collision and visuals. My best guess is that change was necessary to align the multisense with those visuals, but more accurate visuals were then delivered, and that offset was never removed.
Original comment by Steve Peters (Bitbucket: Steven Peters, GitHub: scpeters).
How does a486c18552ad4b69a91a828940b8bf6f27901fcc (branch issue_491
) look?
Original comment by Steve Peters (Bitbucket: Steven Peters, GitHub: scpeters).
I'll make a pull request if you approve.
Original comment by Steve Peters (Bitbucket: Steven Peters, GitHub: scpeters).
Original comment by Steve Peters (Bitbucket: Steven Peters, GitHub: scpeters).
Original comment by Steve Peters (Bitbucket: Steven Peters, GitHub: scpeters).
Original report (archived issue) by Mathew DeDonato (Bitbucket: mdedonato NA).
The multisense is not aligned correctly in the v5 model