Discussing Landmark Definitions and Code Location for Landmark Based Localizer #3945
Closed
SakodaShintaro
started this conversation in
General
Replies: 3 comments
-
There may still be some changes to |
Beta Was this translation helpful? Give feedback.
0 replies
-
Two pull requests were created in connection with this work. |
Beta Was this translation helpful? Give feedback.
0 replies
-
Pull requests have been merged. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
The Localization/Mapping team at TIER IV is currently in the process of implementing landmark based localizer to autoware.universe.
In landmark based localizer, landmarks such as ArUco tags and boards with high LiDAR intensity are detected during localization to calculate the ego vehicle pose.
The landmark is assumed to be a rectangle and is written to the Vector map (lanelet2_map.osm by default) as a 4-vertex Polygon.
For example,
The function that reads landmarks is currently located in autoware.universe/localization/landmark_based_localizer/landmark_parser, but we are thinking of moving it to autoware_common/tmp/lanelet2_extension.
We would like to make time for discussion once here, as there is still room to change or expand the definition of landmarks before moving on.
Should anyone have opinions or suggestions regarding the definitions of the landmarks or the placement of the code, please do not hesitate to leave a comment.
We will wait a week or so and if there are no particular opinions, we will proceed as is.
Beta Was this translation helpful? Give feedback.
All reactions