Localization of a Mobile Robot in Sewer Pipes Using Hough Transform

Robot localization and navigation are ones of the fundamental issues in robotics. The sewer environment contains only a few local features such as manholes, junctions, pipe joints and inlets that are usable as landmarks for localization. In order to navigate an autonomous sewer inspection robot to a given goal (manhole), landmark detection problem needs to be solved. This paper presents a method for landmark detection and self localization by fusion of an IR sensor and a camera. The proposed method is verified with a prototype autonomous sewer inspection robot platform. The result shows 100% of input sewer image provides correct estimation of the distance