You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Use the distance to recognised objects in combination with the data measured by the GPS to better estimate your own position and thus reduce noise.
First make sure that the distance calculation to recognised objects works well enough.
Definition of Done
see description above
Effort Estimate
No response
Testability
No response
Dependencies
No response
The text was updated successfully, but these errors were encountered:
In my opinion, this issue appears to be very difficult. If I understand you correctly, you would like to use dynamic objects for your localization?
Before you use this approach, I would recommend spending at least a small amout of time looking in to point cloud localization methods. If I am correct, high definition maps of the static buildings should be available. You should be able to use these buildings as reference point. However, filtering #459 would be my first approach.
These are only my thoughts, you can do whatever you like.
Yes, we also thought that this issue might be quite difficult and take a lot of time so we are planning on solving the other issues first.
To improve the localization (in particular the estimation of the position) we were planning on using static objects like traffic lights or speed limit signs but buildings as reference points seem to be a good solution as well.
Detailed Description
Use the distance to recognised objects in combination with the data measured by the GPS to better estimate your own position and thus reduce noise.
First make sure that the distance calculation to recognised objects works well enough.
Definition of Done
Effort Estimate
No response
Testability
No response
Dependencies
No response
The text was updated successfully, but these errors were encountered: