Self-learning algorithm for visual recognition and object categorization for autonomous mobile robots

Abstract

In order to execute tasks and to navigate in an environment, an autonomous mobile robot needs a complex visual system to cope with detection, characterization and recognition of places and objects. We are interested here in the development of detection and characterization functions, integrated on a robot. In this paper we consider an approach to the development of categorization systems based on building by a robot of its own semantics, which used only by the robot and is not designed for human perception. © 2012 Springer Science+Business Media B.V.