Vision-based mobile robot localization using natural landmarks

Self-localization in unknown environment is one of the most fundamental tasks for mobile robot. In this paper, a novel localization method for mobile robot working in unknown environment is presented by using scale-invariant natural landmarks. Firstly, corner detector was used to detect the interesting points and get the natural land marks, and then a 3D map of the environment was built. Lastly, the mobile robot localized itself by matching the landmarks in the current frame to a database map. Experiments show that global localization and navigation can be achieved in real time.