Cloud robotics exploits the Internet of things (IoT) for the convergence of shared infrastructure and services for robotics, ranging from using computational resources in the cloud (infrastructure as a service, IaaS) to having a repertoire of algorithms to deploy on a robot on a task-based basis (platform as a service, PaaS). Since its first formulations, different possibilities of cloud robotics have been explored. The DAvinCi project proposed a software framework to parallelize the FastSLAM algorithm to share the computational load among several robots. The RoboEarth project focused on knowledge representation and storage of shared environment information, so that various robots could create semantic maps and perform visual SLAM. Moreover, the ubiquitous network robot platform proposed an integration of robots and sensors by using the cloud for sharing information through an architecture that is highly focused on communications between the various elements. Moreover, the algorithm presented in [6] improved the efficiency of smart factories based on multi-robot services hosted in the cloud, which can run at different speeds and
with different workloads.
