[ Home | Plans | Meetings | Members | Search | Papers | Links | CyberWear | LocoSoft]
External Research Proposal - Situated Computing
University Contact: Professor David May
HP Contact: Jim Bedford-Roberts
Situated Computing concerns the ability of computing devices
to detect, interpret and respond to aspects of the user's local
This is a vast field of computer science that covers sensing
technologies, middleware for interpretation of sensor data,
information infrastructures to provide access to world knowledge
and both novel and situation-enhanced applications [Starner].
The applications in the field offer a tremendous range of end
users benefits. In many cases these benefits are only becoming
instantiable with the advent of ultra-portable computing devices.
Identifying the best commercial opportunities is the real
objective here - becoming overly focused on any of the miriad
individual technologies is the trap. Consequently we would like
to proceed by employing students to cover the wide space of
technology and application development work, leaving university
and HP staff to focus on the more strategic and commercial
Here are some of the questions we would like to address in a
Situated Computing Programme:
- How can we integrate a world of heterogenous sensors,
databases, ultra-portable computing platforms, and
applications [Hull]? This will allow us to understand how
situated computing is likely to unfold in a marketplace
comprising many players with different ambitions.
- What are the most compelling applications of situated
computing? Technical feasibility, infrastructure costs
and end-product costs will all have a bearing on this
- Much of the difficulty of human-computer interaction can
be reduced to the need for users to manually bring the
computer's state into line with their own context of
operation. To what extent can the situated computer's
ability to measure physical context be used to allow it
to develop a shared mental context with their users, and
how much does this ease the human-computer interaction
- How can we develop a general purpose platform that can be
dynamically extended to convert data from as yet
unspecified sensor devices into meaningful event
notifications for as yet unspecified applications?
- What are the social and personal implications of situated
computing (see [Picard] on affective computing)? A
situated computing device has the potential to track the
personal context of the wearer; this raises privacy
concerns [Harper], but at the same time allows the
possibility of brokering between people with shared
interests or backgrounds. At its broadest, situated
computing allows the information world to be overlayed
directly into the real world, with the labelling of
real-world objects and people of interest [Feiner]. What
is the potential when situated computing devices allow
wearers to detect more in their local environment than
just what they can see, hear, touch, taste and smell?
- The ability to respond to things in the proximity
provides situated computers with the potential to make
timely or apparently serendipitous suggestions to wearers
in a proactive manner [Rhodes]. This is in marked
contrast to the reactive pattern of computer behaviour
that occured when the computer user had to initiate
proceedings by providing context-setting information
about a problem to be solved. In what areas will
proactive computers be most useful and should we be
concerned about these devices intimidating their users?
- Although we will develop novel ultra-portable computing
platforms in this programme, our main focus will be on
the interaction and sensing technologies (eg [Hunke] for
face recognition) rather than the platforms themselves.
What mechanisms for HCI work well away from the desk?
What ergonomic issues arise with novel technologies.
Specifically, this programme provides a good opportunity
to evaluate the speech technology research already
underway in the department.
This programme relates to "ultra-portable computing"
work completed in April 1997 by the Interaction Technology
Department (see [Hull1]) and also to a "CyberJacket"
project currently underway with David May and his department (see
web site at http://wear:8080.hpl.hp.com). The CyberJacket Project
is employing three Bristol Computer Science students over the
Summer period in order to build a general purpose, ultra-portable
computer stitched into the lining of an outdoor jacket. This
project is intended to lead directly into the Situated Computing
The general plan for the programme is to perform at least two
test and build cycles for a "CyberJacket" functioning
as a "Bristol Tourist Jacket". Along the way we intend
to address the issues laid out above. The plan can be split into
- Phase 1 - Build first CyberJacket. This will serve as a
proof of concept showing all the diverse components of a
situated computing system integrated into a working
prototype. We will use a traditional PC computer
architecture and off-the-shelf peripheral and interaction
components for the sake of rapid integration. The jacket
will include both GPS and active badge sensors in order
to address "sensor fusion" issues. For this
first CyberJacket we will give priortiy to achieving a
rugged and reliable system, so time will be assigned for
a formal testing programme.
- Phase 2 - Evaluation. We will conduct outdoor field
trials using students wearing CyberJackets. After the
trials we will conduct "unstructured"
interviews in order to gather qualitative information
about the usefulness, functionality, ergonomic operation,
ruggedness and general appeal of the jacket. We will
present our results in a report, including an assessment
of the issues that would arise with the full scale
deployment of situated computing systems in general.
Finally, we will develop a road map indicating the
research and development areas to pursue in the next
generation CyberJacket in order to address the issues
identified so far.
- Phase 3 - Build second generation CyberJacket. This will
serve as a research vehicle for the issues identified in
the road map. Specifically, for this CyberJacket we may
place more emphasis on enabling information exchange
between CyberJacket wearers, eye glass displays, speech
interfaces and power management. In other respects this
CyberJacket may draw on components from the first
- Phase 4 - Evaluation of second CyberJacket. This will
follow a similar pattern to the first evaluation phase,
except that here the field trial may include a team
excercise in order to bring out issues concerning
interaction between CyberJacket wearers.
In summary, objectives and deliverables are as follows:
- Build and refine a proof of concept system showing
integration of situated computing components into a
working prototype. Deliverable: an example CyberJacket.
- Achieve improved understanding of the difficulties to be
overcome before full scale deployment of a situated
computing system is possible. Deliverable: report on
issues raised by the CyberJacket.
- Develop a roadmap for the research and development needed
to allow commercial exploitation of this area.
Deliverable: the roadmap.
Staff and equipment
The breadth of the area means that there are many
mini-projects suitable for undergraduate and master level
projects. At the same time we need to ensure some continuity
between developments and a clear sense of direction. To achieve
this we require an RA working at Bristol University who assumes
ownership of the CyberJacket research platform, control of the
design of critical components, planning and logistics for the
development of CyberJackets and also day to day supervision of
students. In addition we need to set aside funding for project
work including salaries for four students working over each
Summer period. Students will mainly be drawn from the final year
of the Computer Science course, though some post graduate support
may also be desirable.
To equip the programme we need funds for four PCs for
application development and six cyberjackets (three built in each
year of the programme).
The programme will need to last for at least two years
starting FY98. Yearly review will be appropriate. After the two
year period it is likely that we will be in a position to focus
in more specific areas of situated computing in which case a
further ERP will emerge.
||$16k TAC - equipment
|3 CyberJackets and
|4 Summer 98 Students
|3 more CyberJackets and
|4 Summer 99 Students
Start date: FY98.
Internal HP customers: Personal Systems Group
Frequency of project reports: 6 monthly
HP Project Manager: Jim Bedford-Roberts. firstname.lastname@example.org.
[Feiner] S. Feiner, B. MacIntyre, D. Seligmann,
"Knowledge Based Augmented Reality", Comms of the ACM.
[Harper] R. Harper, "Why People Do and Don't Wear Active
Badges", Proc CSCW '96, Kluwer Academic Publisher, pp.
[Hull] R. Hull, P. Neaves and J. Bedford-Roberts,
"Towards Situated Computing", HPL-97-66, International
Symposium on Wearable Computers '97.
[Hunke] M. Hunke, A. Waibel, "Face Locating and Tracking
for Human-Computer Interaction", Proc. of 28th Asimolar
conf. on Signals, Systems and Computers.
[Picard] R. Picard, "Affective Computing", MIT
[Rhodes] B.J. Rhodes, T. Starner, "Rememberance Agent: a
continuously running information retrieval system", Proc.
First Int. Conf. on the Practical Application of Intelligent
Agents and Multi Agent Technology (PAAM '96), pp. 487-495.
[Starner] T. Starner et al., "Wearable Computing and
Augmented Reality", MIT TR355, Nov '95.
The material displayed is
provided 'as is' and is subject to use restrictions.
For problems or questions regarding this web
contact Cliff Randell.
Last updated: January 14, 2000.
ęCopyright Hewlett-Packard 1997-2000.