INTRODUCTION
A recent report from the United Nations indicates that the
average age of the population in developed countries is
increasing. There are more than 22% of elderly people over 65
years old in the world by 2050. They often live alone at home
and have to be autonomous [1]. Therefore, new researches
need to make their live and life in the best possible conditions.
In recent years, home service-oriented robot is hot research
topic for supporting old people. The new robot system design
expects that interaction with friendly appearance [2] [3] will
result in the mental effects of interacting to improve the quality
of life for elderly and children, to reduce loneliness and to
increase security and well-being. Consequently, robot systems
with multimodal human-computer interface provide a
connection between a user and a computer in a natural manner
[4] [5].
Several research groups are exploring the idea of robots
that rely on cloud-computing infrastructure to access vast
amounts of computing power and data [6] [7]. This kind of
“cloud robotic” allows remote compute-intensive takes, such as
video/audio processing and recognition. Some new ones can
even download new skills instantly.
A Google researcher argues that cloud computing could
make robots lighter, smaller, cheaper, and smarter. Moreover,
cloud-enabled robots could offload CPU-heavy takes to remote
servers, relying on smaller and less power–hungry onboard
computers.
This paper presents a human-robot interactive system
design for home senior companion [8] relying on cloud
computing infrastructure, namely “Senior Companion Robot
System (SCRS)”. An example is shown in Fig. 1. Senior needs
a companion while he or she lives alone at home and feels
boring or confusing or needs some information, such as news
and medical information. The accompanied robot will
understand the speech semantics and respond the
corresponding information via SCRS to senior. Every
audio/video processing will be done in SCRS with cloud
information network.