In Humanoids 20 lO, Professor James Kuffner of Carnegie
Mellon University proposed the concept of cloud robotics as
the combination of cloud computing and robotics. Similar to
computers on a network, the robot itself does not need to store
all the knowledge and information and does not need to be able
to perform high-speed computing. The robot can connect to the
server cluster to obtain the required knowledge and information
at the time of need. Cloud robotics is not specific to a robot or a
type of robot. It is the way robots store information and access
a knowledge base. For example, a robot can take photos of the
surrounding environment and upload them on servers. A
computer can search for similar photos and calculate a path
without obstacles for the robot. The corresponding information
and knowledge can be stored centrally and shared with other
robots. Robots can share the knowledge base, thus enhancing
their ability of autonomic learning and decreasing the time a
robot needs to develop or in human terms grow. Based on
cloud computing, robots can have a more powerful data
processing ability. Furthermore, they can submit tasks, such as
natural language processing, image recognition, and speech
recognition, that require high computing power to the remote
servers.