Transcript ppt

A Survey of Socially
Interactive Robots
Terrance Fong, Illah Nourbakhsh,
Kerstin Dautenhahn
Presentation by Dan Hartmann
4/12/2007
dhartman, CS296-3
1
Context - History


The first work in social robotics involved
stigmergy as a model for behavior in insect
colonies
Stigmergy was described to explain how
social insect societies produce complex
behavior patterns, from individuals
performing simple ones.
4/12/2007
dhartman, CS296-3
2
Context - Societies


Insect societies are
anonymous,
homogenous groups.
Many animals form
individual societies,
where each member
forms relationships
and social networks
4/12/2007
dhartman, CS296-3
3
Context – Breazeal’s Four Classes of
Social Robots

Socially Evocative



Relies on the human
tendency to
anthropomorphize



Social Interface

4/12/2007
Socially Receptive
Provides a natural
interface by employing
human-like social cues
Socially Passive but
benefits from interaction
e.g. learning from
demonstration
Sociable

dhartman, CS296-3
Pro-actively engages
with humans to satisfy
internal social aims
4
Context –
Three More Suggested Classes

Socially Situated



Surrounded by a social
environment, they must
be able to distinguish
between social agents
and other objects.
Socially Embedded

Physically connected to a
social environment
requiring at least
rudimentary social
concepts, such as taking
turns.
Socially Intelligent

4/12/2007
Show aspects of human style social intelligence, based on
deep models of human cognition.
dhartman, CS296-3
5
Context – Paper’s Scope


This paper focuses on
"peer-to-peer" humanrobot interaction…
The underlying
assumption is that
humans prefer to
interact with machines
in the same way that
they interact with other
people
4/12/2007

Robots with human
social characteristics
including:
dhartman, CS296-3
6
Methodology – Design Issues

Natural human-robot
interaction



Readable social cues
manifest believable
behavior, establish
appropriate social
expectations

Must send signals to
the human to provide
social feedback.
Real-time performance

4/12/2007
Must operate at human
interaction rates
dhartman, CS296-3
7
Methodology –
Embodiment
“That which
establishes a basis for
structural coupling by
creating the potential
for mutual perturbation
between system and
environment"
- Authors’ definition
4/12/2007
dhartman, CS296-3
8
Methodology - Embodiment

Morphology


Appearance biases interaction (e.g. a robot dog will be
treated differently than an anthropomorphic robot)
Design Considerations


4/12/2007
Needs enough humanness
for user comfort
Needs enough robot-ness to
prevent false expectations of
the robot's capabilities
dhartman, CS296-3
9
Methodology - Embodiment

Anthropomorphic


Many argue that to interact
socially with people a robot
should resemble a human
Zoomorphic


Caricatured



Realism is not necessarily
needed for believability.

Functional

4/12/2007
The embodiment should reflect
the tasks it must perform.
dhartman, CS296-3
Most common are "pet"
type robots
Human-creature
relationships are simpler
than human-human
relationships
Easier to avoid the
"uncanny valley“ in
previous slide
10
Methodology –
Human-Oriented Perception



To interact with humans in the real world,
social robots must perceive the world the
same way that humans do
In particular, they must be able to track
human features and interpret human
communication
Similar perception may require similar
sensing
4/12/2007
dhartman, CS296-3
11
Methodology –
Human-Oriented Perception

Each of these tasks is mentioned and
references papers for in depth work.




4/12/2007
People Tracking
Speech Recognition
Gesture Recognition
Facial Perception
dhartman, CS296-3
12
Topics That I Am Skipping
for Time and Relevance
 Dialogue
 Personality
 Emotion
4/12/2007
 User
Modeling
 Socially Situated
Learning
 Intentionality
dhartman, CS296-3
13
Discussion –
Attitudes Towards Robots

Khan describes a survey to investigate
people’s attitudes towards intelligent
service robots. Two significant findings
were:


4/12/2007
A robot with machine-like appearance, serious
personality, and round shape is preferred
Verbal communication using a human-like voice
is highly desired.
dhartman, CS296-3
14
Discussion – Field Studies

Scheeff et al. conducted two studies to
observe how a range of people interact with
a creature-like social robot.


4/12/2007
Children were observed to be more engaged
than adults.
A friendly robot personality was reported to have
prompted qualitatively better interactionthan an
angry personality.
dhartman, CS296-3
15
Discussion –
One Last Point for Perspective

Paraphrasing Wood,the authors say:

4/12/2007
“Humans and robots must be able to coordinate their
actions so that they interact productively with each other.
It is not appropriate (or even necessary) to make the robot
as socially competent as possible. Rather, it is more
important that the robot be compatible with the human’s
needs, that it matches application requirements; that it be
understandable and believable, and that it provide the
interactional support the human expects.
dhartman, CS296-3
16