search

SteveMacenski / slam_toolbox

Slam Toolbox for lifelong mapping and localization in potentially massive maps with ROS
GNU Lesser General Public License v2.1
1.68k stars 525 forks source link

Possible to do re-localization in the life-long-slam mode? #522

Closed xingjl6280 closed 2 years ago

xingjl6280 commented 2 years ago

Feature description

I want to try the life-long slam mode in the real world. The problem is the robot might be turned on anywhere, therefore need to manually re-localization. Is that possible to let the life-long node subscribe and process initialpose just like the pure localization node?

SteveMacenski commented 2 years ago

Lifelong SLAM mode is a not super functional prototype under experimental, so in all honesty this is kind of a moot question because that mode is not really ready for prime-time use (hence why its in experimental).

But conceptually, you can set the initial pose when you deserialize your slam graph to provide it with the initial location, but you cannot manually re-localize after that point (nor could you with any other slam mode).