The SLAM (Simultaneous Location and Mapping) algorithm is how the TurtleBot recognizes the area it is in. The SLAM algorithm creates a top-down map of the area by physically moving around the area (controlled by some manual mechanism) and will use various sensors inside the TurtleBot's Kinect camera to see how close certain things are to it and keep track of both where it is and what it has seen so far. A demonstration of how this process looks like can be seen in the following video.
From this video, you can see that the TurtleBot will take a look at it's surrounding and map where it has gone so far and where it can and can't go to.
The TurtleBot can use a pre-made map to move on it own around an area. Once a map is created, the map-maker can place marker to which the TurtleBot can freely move to. If something about the area has changed since the map's creation, the sensors on the TurtleBot will be able to safely prevent the robot from colliding with objects. A video of that process can be seen below.
This is a video of a TurtleBot being told to go in various directions. The TurtleBot uses it's sensors to be able to know not to collide with objects and where it has already been to.
Design from this website taken from CSCI322 Client/Server Project.