So far most of what we had was theory. Now we wil see how autonomous navigation is implemented in ROS. More specifically, the ROS Navigation stack.

• The ROS Navigation Stack is meant for 2D maps, square or circular robots with a holonomic drive, and a planar laser scanner, all of which a Turtlebot has. It uses odometry, sensor data, and a goal pose to give safe velocity commands.
• The node “move_base” is where all the magic happens in the ROS Navigation Stack.
• Uses a global and local planner to accomplish the navigation goal.
• Manages communication within the navigation stack.
• Sensor information is gathered (sensor sources node), then put into perspective (sensor transformations node), then combined with an estimate of the robots position based off of its starting position (odometry source node). This information is published so that “move_base” can calculate the trajectory and pass on velocity commands (through the base controller node).