Sentient Computing
Andy Hopper
Presenter : Youn Do Lee
Oct 31, 2005
Why this presentation
There have been many presentations
about location systems
Active bat, Active badge, Cricket,
Radfinder, etc.
This presentation is about issues beyond
the location systems
Refining location information
What to do with location information
Contents
INTRODUCTION
REASERCH ISSUES
Location sensing
Spatial monitoring
Data distribution
Applications
OBSERVATION
Sentient Computing?
Sentient computing is
The proposition that applications can be
made more responsive and useful by
observing and reacting to the physical world
A form of ubiquitous computing which uses
sensors to perceive its environment and react
accordingly
Sentient computing project
AT & T Laboratories Cambridge
Active bats
Follow-me systems
Now defunct
Reference
Mike Addlesee , Rupert Curwen , Steve
Hodges , Joe Newman , Pete Steggles , Andy
Ward , Andy Hopper, Implementing a Sentient
Computing System, IEEE Computer Magazine,
v.34 n.8, p.50-56, August 2001
Location Sensing
Categorising the concept of location
Containment
Active badge, Cricket
Proximity
Bluetooth
Co-ordinate
GPS, Active bat
Active badge
Active bat
Spatial monitoring
Location systems provide raw spatial
facts about objects
But location-aware applications need
more than raw spatial data
They need spatial relationships between
objects that are significant
How to decide whether a spatial
relationship is significant
Spatial monitoring (cont’d)
One possible approach
Operating on the basis of zones of
containment surrounding objects
Person X is “holding” Person X can be “seen” by camera B
Keyboard K but not by camera A
Data distribution
Beyond location-publishing application
An automatic control of the digital
environment without user intervention
For example
The personal desktop follows the user to any
nearby device
To archive this, A platform for connecting and
displaying information on all these devices
are needed in addition to location information
Data distribution (cont’d)
VNC (Virtual Network Computer)
A simple device independent protocol
The viewer has no state, and simply displays
information graphically
The connection from
viewer to server is
also stateless, just
keystrokes and
pointer clicks
Applications
Opening and closing doors automatically
A textual indication of where someone is, how
fast the are moving, how long they have been
there
Showing the local context, including who and
what else is nearby
Personalization by teleporting VNC desktops
Surveillance applications which the selection of
a particular camera is based on spatial data
Contents
INTRODUCTION
REASERCH ISSUES
Location sensing
Spatial monitoring
Data distribution
Applications
OBSERVATION
Observation
Attempts at automatic control without user
intervention have not proved enduring
Automatically teleporting to the nearest screen
Automatic routing of phone calls
Once more than a simple inference is attempted
it seems that hit a brick wall
The sighting of three or more Active Badges in a
single space => “meeting” ?
Observation (cont’d)
One potential research direction is to provide
much more feedback to the user
Visual and aural feedback with perhaps every nearby
wall being used display may be one approach
The user can interact in a much more informed way
and help guide any decision-making process
Perhaps a way to make progress beyond the
engineering level is to imagine a “perfect”
sensing system with full coverage of the
environment
Thank you, any question?