Localization

As just mentioned, global localization is the problem of finding out where a robot is relative to a map (i.e. align the robot's local coordinate system with the global coordinate system of the map). This problem is as important as being able to build a good map of the environment. No matter how good the map is, it will be of no use if we are not able to localize the robot on it. Conversely, even if we know how to localize the robot with high precision, that will be useless if there is no good map available on where to localize it. Moreover, the accuracy of a metric map depends highly on the alignment of the robot with its map. If we are not able to localize the robot, the resulting maps are too erroneous to be of practical use. As seen, these two problems are closely related, and most of the mapping approaches try to address both problems at the same time, in what is known as simultaneous localization and mapping (SLAM).