Project-MANAS / slam_gmapping

Slam Gmapping for ROS2
123 stars 37 forks source link
gmapping-slam localization mapping robotics ros2 slam

SLAM_GMAPPING

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. 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.

Launch:

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.