A mobile delivery robot has at least one memory component containing at least map data; at least two cameras adapted to take visual images; and at least one processing component. The at least one processing component is adapted to at least extract straight lines from the visual images taken by the at least two cameras and compare them to the map data to at least localize the robot. The mobile robot employs a localization method which involves taking visual images with at least two cameras; extracting straight lines from the individual visual images with at least one processing component; comparing the extracted features with existing map data; and outputting a location hypothesis based on said comparison.