Seamless mobile robot localization service framework for integrated localization systems

Abstract

According to its importance as one of the most basic preconditions for autonomy of mobile robots, efforts to provide global localization continuously without loss have been advanced so far. With current trends toward indoor service robots like security robots, cleaning robots, and home service robots, not only outdoor but also indoor global localization is demanded. However, despite both outdoor and indoor location systems have achieved reasonable accuracy of localization, still none of existing localization systems guarantees seamless and transparent global localization services regardless of indoor or outdoor due to different requirements. To address seamless and transparent global localization services irrespective of whatever underlying technologies are used, we propose Integrated Localization Service Framework for mobile robots. The framework provides abstraction of global localization services for simple development and management of mobile robots. Our key considerations for the proposed framework are seamlessness by applying new localization algorithm and transparency by designing new architecture of fusion. The localization algorithm provides location fix even in exceptional case like handoff, and the fusion method allows utilization of measurement data from anchors using different localization technologies for one location fix, differently from other systems only can localize with more than three measurement data based on the same localization technology. We also implement an integrated localization system by embodying the proposed framework and, designing and integrating 4 localization modules based on IEEE802.15.4, IEEE802.15.4a, Wi-Fi, RFID, and GPS via a localization main board. To verify the system, we evaluate the localization accuracy and seamlessness by experiments. As a result, the experimental results show the proposed framework achieves reasonable localization accuracy compared to other existing localization systems as addressing se...