Real Time Mobile Robot Localization by Using a Laser and a Geometric Map

Abstract Mobile robot localization is the task of calculating the position and orientation of the robot through external sensorial information. There are two main approaches: landmarks detection and the use of an a priori map of the environment. This paper follows the second approach. The sensorial system is a laser beam projector and a camera. From all the points detected in the environment the straight lines are extracted, then they are matched with the map of the environment. The algorithm is based on the Extended Kalman Filter and has been tested in real situations.