SLAM(Simultaneous Localization and Mapping) is the computational problem of constructing or updating a map of an unknown environment while simultaneously keeping track of an agent's location within it.
This contains package openslam_gmapping
and slam_gmapping
which is a ROS2 wrapper for OpenSlam's Gmapping. The wrapper has been successfully tested with Eloquent Elusor
and Foxy Fitzroy
. Using slam_gmapping, you can create a 2-D occupancy grid map (like a building floorplan) from laser and pose data collected by a mobile robot.
ros2 launch slam_gmapping slam_gmapping.launch.py
The node slam_gmapping subscribes to sensor_msgs/LaserScan on ros2 topic scan
. It also expects appropriate TF to be available.
It publishes the nav_msgs/OccupancyGrid on map
.
Map Meta Data and Entropy is published on map_metadata
and entropy
respectively.