An environmental visual features based navigation for mobile robot in a corridor environment

This paper presents a navigation method that enables an autonomous mobile robot to localize itself and identify its own orientation in order to follow a path in the environment. Both tasks use the same recognition method. The method is based on features data provided through an image captured by a single camera, which is trained by a neural network. No precise and accurate measurement is used in the proposed navigation method, but the experiment results showed that the proposed method is competent to work in actual environments.