Navigation and localization are two important issues which need to be addressed in order to let robots work in the human living environment. We believe that it is not always necessary to build a robot system with a precise and accurate navigation in some situations. For example, those cases where a robot is programmed for guiding way in a building environment or fulfilling a delivery task. With this regard, we propose a navigation method for the autonomous mobile robot where the robot will identify its own position and orientation robustly based on visual features in the environment without any complicated techniques. In our method, the robot takes advantage of the localization process to navigate from a point to a given destination without losing the correct path.