USING COLOR IMAGE FEATURES IN DISCRETE SELF-LOCALIZATION OF A MOBILE ROBOT

Natural landmarks are assumed to exist in the environment. Global color image features are extracted from sensor data to feed the robot’s self-localization approach. The color features correspond to natural landmarks, that are learned by the navigation sub-system. During the localization process, which is a Bayes filtering of a Markov environment, the posterior probability density over possible discrete robot locations (the belief) is recursively computed. The approach was tested to provide robust results under varying scene brightness conditions and small measurement errors.