Robot Mapping and Navigation by Fusing Sensory Information

Abstract

environment. The acquired knowledge through learning process is used to build an internal representation. Knowledge differs from information in that it is structured in long-term memory and it is the outcome of learning. In order to enable an autonomous mobile robot to navigate in unknown or changing environment and to update in real-time the existing knowledge of robot’s surroundings, it is important to have an adaptable representation of such knowledge and maintain a dynamic model of its environment. Navigation in unknown or partially unknown environments remains one of the biggest challenges in today's autonomous mobile robots. Mobile robot dynamic navigation, perception, modeling, localization, and mapping robot’s environment have been central research topics in the field of developing robust and reliable navigation approach for autonomous mobile robots. To efficiently carry out complex missions in indoor environments, autonomous mobile robots must be able to acquire and maintain models of their environments. Robotic mapping addresses the problem of acquiring spatial models of physical environments through mobile robots and it is generally regarded as one of the most important problems in the pursuit of building truly autonomous mobile robots. Acquiring and mapping unstructured, dynamic, or large-scale environments remains largely an open research problem. (Habib & Yuta, 1988; Kuipers & Byun, 1991; Thrun & Bucken, 1996; Murphy, 2000; Thrun, 2002). There are many factors imposing practical limitations on a robot’s ability to learn and use accurate models. The availability of efficient mapping systems to produce accurate representations ofinitially unknown environments is undoubtedly one of the main requirements for autonomous mobile robots. A key component of this task is the robot’s ability to ascertain its location in the partially explored map or to determine that it has entered new territory. Accurate localization is a prerequisite for building a good map, and having an accurate map is essential for good localization (Se et al., 2002;Choset,2001 ). All robots,which do not use pre-placed landmarks or GPS must employ a localization algorithm while mapping an unknown space. Therefore, accurate simultaneous localization and mapping (SLAM) represents a critical factor for