It is very important for a mobile robot to estimate its current position. With precise information about the current position, the mobile robot can do path-planning successfully. In this thesis, a position estimation method using one color image is presented. The mobile robot(K2A) takes an image of a corridor and searches for the door and pillar, which are the given landmarks. Color information is used to distinguish the landmarks. In order to represent the presence of the landmarks, image mode is defined. This method adopts Kullback information distance. If the landmarks are detected, with the color information, the mobile robot identifies the vertical line of the pillar and its crossing point with the floor. After finding the crossing point, the mobile robot calculates the distance from its current position to the crossing point and a primitive robot navigation is performed. With the help of color information, it is quite simple to implement the searching algorithm to find the landmarks and desired vertical line and it takes 4 seconds to carry out the procedure.
