ROS 2 | TurtleBot3 SLAM and Autonomous Navigation with Sim2Real Transfer [Tutorial]

This video demonstrates SLAM and autonomous navigation of a TurtleBot3 (a differential-drive robot) in Gazebo Simulator as well as real world using Robot Operating System 2 (ROS 2).

During the simultaneous localization and mapping (SLAM) phase, the robot starts off by building a 2D binary occupancy grid map of the environment using Cartographer SLAM algorithm, which is then saved for navigational purposes.

During the autonomous navigation phase, the robot loads the static map of the environment and takes an initial pose estimate for localization with respect to that map. Later, once a desired goal pose is given (i.e. published), the robot uses Nav2 stack to plan a global path (using the knowledge of the static map) from its initial pose estimate to the goal pose and also dynamically re-plans a local trajectory (using the knowledge of dynamic local costmap based on real-time LIDAR scans) from its current pose (determined using particle-filter based localization) to the goal pose, which is then tracked in real-time using motion control algorithm. The recursive perception-planning-control loop is executed until the robot is within a specified tolerance of the desired goal pose (in terms of both position as well as orientation).

TurtleBot3 Packages: https://github.com/Tinker-Twins/Turtl...
ROS 2 Installation:    • ROS 2 | ROS 2 Foxy Installation [Tuto...  

------------------------
Specifications:
------------------------
OS: Ubuntu 20.04
ROS 2: ROS 2 Foxy