ABSTRACT
This chapter presents a cloud-based architecture to achieve large-scale localization of mobile robots in outdoor environments, taking advantage of the powerful computation, storage, and other shared resources of the cloud. Cloud robotics, a concept first introduced by Kuffner at Google, provides a very promising solution to the problems faced by large-scale or long-term autonomous robots. The chapter introduces a different cloud-based localization architecture using outsourced road network maps and reference images in the cloud (CLAOR) to achieve real-time autonomous navigation of a mobile robot in large-scale outdoor environments. CLAOR has two phases: offline and online. The offline phase consists of two parts: the extraction of road network information and the construction of reference images. The former extracts new road networks and updates the existing road network map stored in the cloud. The online phase is for mobile robot localization based on the cloud. The chapter discusses the particle-filter-based localization algorithm designed on the robot to estimate its position.
