| ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
Seems like AMCL does a really bad job of localizing in the map. I'm using the global_localization service to distribute the particles across the map initially, but the results are things like this:
when in fact the robot is staring at a corner like this:
Thoughts on what I'm missing here? Or what I need to look into to improve my results?
If you are just interested in the localization and it doesn't matter if it is global localization, then you can give it starting pose through the "2D Pose Estimate" button in Rviz. Then it will try to localize with respect to this initial point.
The issue with AMCL is that with the global localization you cannot be sure it converges to the correct localization due to aliasing and the probabilistic nature of the particle filter.
Asked: 2016-08-19 16:32:35 -0500
Seen: 959 times
Last updated: Aug 22 '16
ROS Answers is licensed under Creative Commons Attribution 3.0 Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.
