to interact with people. Many of these systems target different application domains
such as computer interfaces, Web agents, synthetic characters for entertainment, or
robots for physical labor. In general, these systems can be either embodied (the
human interacts with a robot or an animated avatar) or disembodied (the human
interacts through speech or text entered at a keyboard). The embodied systems have
the advantage of sending para-linguistic communication signals to a person, such as
gesture, facialexpression, intonation, gaze direction, or body posture. These
embodied and expressive cues can be used to complement or enhance the agent’s
message. At times, para-linguistic cues carry the message on their own, such as
emotive facialexpressions or gestures. These embodied systems must also address the
issue of sensing the human, often focusing on perceiving the human’s embodied
social cues. Hence, the perceptual problem for these systems is more challenging thanthat of disembodied systems. This section highlights a few embodied systems, both
animated and robotic.
2.1. Embodied conversation agents
There are a number of graphics-based systems that combine naturallanguage with
an embodied avatar. The focus is on natural, conversational discourse accompanied
by gesture, facialexpression, and so forth (Cassell, 1999). In some applications, the
human uses these systems to perform a task. One of the most advanced systems in
this respect is Rea from the Media Lab at MIT (Cassell et al., 2000). Rea is a
synthetic real-estate agent, situated in a virtual world, that people can query about
buying property. The system communicates through speech, intonation, gaze
direction, gesture, and facial expression. It senses the location of people in the room
and recognizes a few simple gestures. Another significant application area is tutoring
systems where the agent helps a person learn how to perform a task. An advanced
pedagogicalsystem is Steve, developed at USC (Rickeland Johnson, 2000). The
human is immersed in virtualreality to interact with the avatar. It supports domain-
independent capabilities to support task-oriented dialogs in three-dimensional (3D)
virtual worlds. For instance, Steve trains people how to operate a variety of
equipment on a virtualship and guides them through the ship to show them where
the equipment is located. Sometimes, the task could simply be to communicate with
others in a virtualspace, a sort of animated ‘‘chatroom’’ with embodied avatars
(Vilhjalmsson and Cassell, 1998). There are a number of graphicalsystems where the
avatar predominantly consists of a face with minimal to no body. In Takeuchi and
Nagao (1993), for instance, the use of an expressive graphicalface to accompany
dialog is explored. They found that the facial component was good for initiating new
users to the system, but its benefit was not as pronounced over time. Also of note in
this issue is the paper on Greta by de Rosis et al. that describes how Greta’s verbal
and non-verbalsignals are synchronized and animated in 3D.
2.2. Human-friendly robots
The ability to interact with people in the human environment has been a recent
motivator of the humanoid robotics community and the service robotics community.
For systems such as these, safety and minimizing impact on human living spaces are
important issues, as well as the issues of performance and ease of use. For example,
the MOVAID system (Dario and Susani, 1996) and a similar project at Vanderbilt
University (Kawamura et al., 1996) focus on providing assistance to the elderly or to
the disabled. In a more educational setting, a number of mobile museum tour guide
robots are employed around the world such as Sage from the University of 情感和社交的人形机器人英文文献和中文翻译(2):http://www.751com.cn/fanyi/lunwen_12732.html