UBC Publications

UBC Publications

UBC Publications

Focus 1993

Item Metadata


JSON: focus-1.0115152.json
JSON-LD: focus-1.0115152-ld.json
RDF/XML (Pretty): focus-1.0115152-rdf.xml
RDF/JSON: focus-1.0115152-rdf.json
Turtle: focus-1.0115152-turtle.txt
N-Triples: focus-1.0115152-rdf-ntriples.txt
Original Record: focus-1.0115152-source.json
Full Text

Full Text

Array Vol. 4, No. 1
Spring 1993
A look at how the work of
CICSR members will impact
future products and processes
■ We are the future. This has been the
motto of the UBC Centre for Integrated
Computer Systems Research (CICSR) for
years. The focus of this issue is to illustrate
the many ways CICSR lives up to its
CICSR was formed to encourage interdisciplinary, industry-driven research in
Computer Science, Electrical Engineering
and Mechanical Engineering. As these
fields evolve, the amount of cross-
... continued on page 3
Space Station research involves researchers from
all CICSR disciplines.
Artist's rendering courtesy MacDonald Dettwiler
Design Optimization
page 2
A better way to
handle errors
page 3
Smarter machines
page 4
Outreach philosophy
page 5
ASI promotes university-
industry collaboration
page 6
Vision of the future
page 7
page 8 The Shape of Things to Come
■ This newsletter is the last to be
produced out of our current office in the
CEME Building. This summer, we will
I move into the new CICSR/CS Building,
| currently under construction on Main
Mall. The next few months will be an
exciting, busy time for all CICSR and
Computer Science Department personnel,
as we prepare for this move. We are
planning a special "housewarming party"
for the middle of October which will
include the official opening ceremonies,
an Open House, and a symposium to
honour the 25th anniversary of the
Department of Computer Science. Details
will appear in the next issue.
We've just concluded our Distinguished
Lecture Series for 92/93 on System
Design and Software Engineering. Next
year's DLS will involve Parallel and
Distributed Systems, and the list of
speakers is given in this newsletter.
Notice that the talks will be held in our
new building! A second series will be
initiated this Fall, called the CICSR
Faculty Forum. It will feature six CICSR
faculty members describing the research
work of their laboratories.
This issue of our newsletter highlights the
work of several more CICSR researchers,
and how their work will impact future
products and processes. The development
of these new technologies is a difficult,
often frustrating experience, but crucial to
our society's future well-being.
One key element of this development is
the link between university research and
industrial development. This interface is
the focus of the work of the B.C. Advanced Systems Institute. Its new Director, Brent Sauder, is highlighted in this
newsletter. As I'm writing this, we've just
completed a very successful ASI Grad
Student Presentation Day, where 60 grad
students from the three universities
presented their research work to local
industry and academia. It was an excellent
example of the kind of interaction
between students, faculty and industry so
necessary if we are to succeed in transferring technology to the marketplace. ■
Dr. James Varah, CICSR Director
■ The theory behind CICSR researcher
Mohamed Gadala's field, design optimization and non-linear finite element applications, has been around for a long time. But
it is only recently that real-world applications of the research have become practical. That's because it takes a tremendous
amount of computing power and memory
to grind through the calculations it takes to
come out with an optimal design. Both the
software and hardware for design optimization and non-linear finite element
analysis have only recently become
available for practical research and
industry use.
Gadala is at work setting up a lab in the
Department of Mechanical Engineering
with the required software and hardware to
perform his work. He is focused on design
optimization using finite elements, and the
general application of finite elements to
non-linear problems.
Gadala says the lab is almost complete.
The initial investment in hardware and
software is considerable. "I'm very
fortunate. Most of the equipment I need is
already here," said Gadala. The computers
required to perform design optimization
calculations have to be several times as
powerful as a typical high-end workstation.
The problems involve imaging, graphic
display, solving problems with a large
system of equations, and they must be
iterative solutions. Before the advent of
today's powerful workstations, this type of
research couldn't be done in any practical
way without a mainframe computer, said
Gadala. Fortunately the tools are now more
Potential applications for this research are
widespread, from automotive and aerospace industries to applications for nuclear
Mohamed Gadala's research field of design
optimization and non-linear finite element
applications will help form the shapes of the
products of the future.
power generation. In fact, Gadala worked
for years with Atomic Energy of Canada
and Ontario Hydro applying his knowledge
to nuclear power stations. After that, he
worked in Michigan with a number of
companies connected to the car manufacturing industry.
Design optimization can be applied to any
system or part that it makes sense to
optimize, either because that part is
manufactured by the millions so that small
cost-savings or improvements make big
differences, or because a system is
extremely valuable. "Normally design
optimization problems are at the higher
end of the design engineering phase," said
Gadala. "For example, I might take a
specific component of a car and try to
improve it in terms of reducing its weight
or cost."
In Europe, design optimization is more
commonly practised than in North
America, and it is often employed at the
initial phase of the design. Gadala said the
concept is just starting to catch on in North
America, though it's often applied after the
fact to improve an existing product that is
simply not competitive. "We are working
to make local engineers and companies
more aware of the fields and its benefits."
The result will be superior-quality products
and systems. Automotive and aerospace
industries provide some real-world
examples of the benefits of design optimization. German car manufacturers such as
Daimler Benz and BMW have embraced
the concept, and their products are sought
after and associated with high quality.
Design optimization ultimately changes the
look and shape of products it is applied to.
Just look at how automobile shapes have
evolved over the years from older,
squared-off cars to today's sleek, aerodynamic silhouettes.
Design optimization is just starting to
become more widely taught and more
widely available. UBC will be offering a
graduate course in the area, taught by
Gadala, which for the first time will enable
student to get beyond theory to the
practical side of the field. "The field is
starting to pick up," said Gadala, and as it
does, it will change the forms of the
products and systems we'll be using in the
future — and change them for the better. ■ THE FUTURE ... continued from cover
disciplinary knowledge required is
increasing. The line between hardware and
software, for example, is blurring. And the
need to marry computer and mechanical
systems seamlessly for industrial applications is increasing.
A good example of the need for CICSR's
approach is the work required to develop
robots for the Space Station. While the
story in this newsletter focuses on the
vision system being developed for the
Space Station by David Lowe of the
Department of Computer Science,
researchers from all CICSR disciplines are
working on aspects of the Space Station in
conjunction with a small company called
Kinetic Sciences.
Lowe notes that vision systems make
good CICSR projects because by them
selves, they are not all that useful. Their
practicality lies in their integration with
computer and mechanical systems to
make the industrial robots and mechanical
products of the future.
Mohamed Gadala's work in design
optimization and non-linear finite element
applications (described in this issue) also
has implications for the future. To date,
the knowledge has been applied mainly in
the automotive and aerospace industries to
optimize design, manufacturability, cost
and performance of various parts. But
now, the tremendous computing power
required for design optimization is
becoming more accessible, so work like
Gadala's will become more and more
Ian Yellowley, an Emeritus Fellow of the
B.C. Advanced Systems Institute and
CICSR member has developed a new
control architecture for machine tools that
has major commercial potential. Well-
funded teams worldwide are working on
the problem, but so far Yellowley and the
UBC team have the most elegant solution.
Also covered in this issue is the work of
Computer Science Department Head
Klawe in two areas: optimal alphabetic
binary search trees, and educational video
games for teaching math. We talk with
B.C. Advanced Systems Institute executive
director Brent Sauder, and Electrical
Engineering professor, Samir Kallel, who
is working on algorithms to correct errors
in the transmission of data over virtually
any medium. All of these CICSR members, and in fact the entire CICSR faculty,
are working on projects of an applied
nature that will change and improve the
products and processes of the future. ■
A Better Way to Handle Errors
■ Samir Kallel, a CICSR member and
holder of a New Faculty Award from the
B.C. Advanced Systems Institute, is
working on ways to control errors in the
transmission of data over virtually any
medium from satellite to radio to phone
lines. According to Kallel, the possibility
for error in transmitting data always exists.
His work involves reducing or controlling
errors so reliable information is transmitted to communications systems users.
One method of dealing with errors is called
forward error correction. This method was
used only in very high-end applications,
such as in military and space communications, but is now far more widespread. In
forward error correction, the transmitter
sends the desired data, plus some built-in
redundancy. At the receiving end, the
redundancy is exploited to correct any
transmission errors that have occurred.
An analogy might be that if you're talking
to someone across a room, and they have
trouble hearing you, you add more
information to your message so the
receiver can better interpret what you're
saying. A simple response to this might be:
"Why not just speak louder?" In the
communications world, speaking louder
would be equivalent to adding power.
However, one of the purposes of forward
error correction is to enable legible
transmissions using less power. This can
significantly reduce the cost of a transmission system. Forward error correction can
also be used to maximize the rate of
transmission and still have usable information at the other end.
Samir Kallel is developing error-correction methods for data transmission over virtually any medium.
The key to forward error correction is to
choose the best code and a good way of
decoding it at the receiving end. This
technology is being applied in an increasing range of applications, and Kallel is
working on designs for many of these.
He is also working on another way of
dealing with errors, which can be applied
to systems where the receiving end can tell
the transmitter if and when it is receiving a
message with errors. Kallel is working on
a system that will detect errors and retransmit. Extra code is sent using this
method as well, but it is used only for
error-detection purposes.
A combination of the two error-handling
methods is also in the works. This hybrid
system will detect errors, and then rather
than re-transmitting the same message, will
transmit redundancy similar to that used in
forward error correction. Kallel is also
studying an incremental redundancy
scheme that will send more and more
redundancy as needed to decipher the
message at the receiving end. "It's better
than sending the same message again and
again," said Kallel.
He is currently working on a Science
Council of B.C. project to design and
implement a mobile data link protocol.
Both MPR Teltech and Motorola are
interested in the work, said Kallel. He adds
that his ASI New Faculty Award enables
him to more easily pursue work with local
industry. "Ultimately, I'd like to incorporate our ideas into their products." ■ Smarter Production Machines
UBC team winning the race to develop this coveted technology
Smarter production machines have long
been the goal of researchers worldwide.
However the new system architecture for
machine tool control, designed by a team
at UBC, is one of the very first ready for
Ian Yellowley, an Emeritus Fellow of ASI,
and CICSR member, has headed the
development of an open control architecture which has been running an industrial
scale, (25KW spindle), three-axis lathe in
the Manufacturing Engineering Laboratory
at UBC for the last two years. Investors
are in place to commercialize the research
once patent protection is obtained.
According to Yellowley, word is imminent
on whether a first, broad ranging patent on
the technology will be approved.
There is worldwide interest in this
technology and numerous well-funded
consortia have sprung up, particularly in
the U.S., to pursue it. Two years ago,
when the first patent application was
submitted, the UBC group felt that it was
well ahead of the competition. At this
point, given the relative funding levels
available to the various groups, the UBC
team finds itself in a situation which is all
too common in Canada, according to
Yellowley. "It's a bit like being in a soap
box powered by an elastic band, with all
these turbo Porsches roaring up in your
rear view mirror."
However, in conjunction with the UBC
Industry Liaison Office, he is working to
construct a fence around the technology
through the patenting process. At least two
more applications are expected following
the feedback on the first application. In the
meantime, research is ongoing.
General Motors of Canada recently
donated a five-axis GMF robot to the lab
and Yellowley, with graduate student Rudi
Seethaller, is working to apply the UBC
architecture to this machine. In general,
what he and so many others are trying to
accomplish is to design advanced machine
controls with considerable on-board
intelligence. The result will allow expensive production machines to operate more
quickly and with less error, ultimately
reducing the processing cost of products
made by these advanced machines.
A specific use of this technology is in
minimizing the cost of making parts using
machine tools. The UBC control architecture features parallel computation of
process parameters and allows integration
of process and geometry in a manner
which is well beyond that currently
available. The capabilities of the system
have been demonstrated in work by
graduate student Ramin Ardekani.
The system does not simply measure basic
parameters such as force or torque — it
goes beyond these to calculate machining
parameters such as depth of cut and mean
chip thickness so that tool breakage may
be avoided in complex contouring operations. It then examines the whole process
and drives the system towards optimality
while always avoiding constraints.
"Our system will sense a problem and
deliver a message to the axis processors in
no more than 1 millisecond? and usually
much faster," said Yellowley. The system
is useful both for optimization and for the
extra protection afforded to the operator,
machine tool and component being
In order to get to the point which the UBC
architecture has reached, the team had to
start from scratch, building controllers
which could run faster without the need for
expensive hardware. This involved
considerable study of the influence of
architecture and algorithms on the hardware requirements. In addition research
engineer Phillip Pottier undertook the
development of the original axis processor
to allow the demonstration of the ideas
before such high performance
microcontroller based cards were available. (The UBC team is now able to
purchase the boards and modify them to
meet their needs, but this was not the case
two years ago. The board development
constituted quite a departure for a group of
mechanical engineers!)
" The major issue in robot control is the
need for high speed interpolation with low
following error. We have had to develop
new improved techniques to control error
under such circumstances," said
Yellowley. These algorithms run in
parallel with force or joint torque constraint algorithms. The machine then
knows its own limitations, so it is impossible to push it too fast.
The UBC group is now working with a
local company to put the technology on a
numerically controlled (CNC) milling
machine. In a parallel development, the
group is embedding a full CAD/CAM
interface into the controller with support
from Autodesk and NC Micro Systems.
The system runs on a second 486 system
sitting on the same bus as the CNC master.
This will allow the programming editing
and downloading of design information
while the machine tool is in operation.
Although it wasn't their original goal,
Yellowley said that the interest in commer-
... continued on next page
Yellowley and his team are winning the race to develop smarter production machines for commercial uses. YELLOWLEY ... continued from page 4
cialization is very high. The National
Research Council has standardized on the
UBC architecture for their five-axis
machining center in Ottawa, and is
pursuing commercial Canadian companies
to use the system as a test bed for the
development of new machine tools.
For now, the UBC development appears to
be well placed to outperform the large
well-funded consortia that are researching
in the same area. UBC's main advantage is
that the team has been able to spend
considerable time examining the fundamental problems, and developing simple
but elegant algorithms which allow
significant cost savings in actual hardware.
Yellowley says it is likely that the UBC
architecture will cost about one half of
other systems under development with the
same projected capabilities. ■
Klawe's Outreach Philosophy
Computer Science Department Head Maria Klawe believes
university research should benefit society at large
■ Computer Science Department
Head Maria Klawe believes university
researchers should work together for
the benefit of society at large. "In our
departments, we have so many
talented people. Our core research is
important, but it's also important to
pay attention to the problems of
society and address them if possible.
It's part of our outreach philosophy,"
she said.
A good example of this philosophy in
action is her work with various
members of the UBC research
community in designing educational
video games to make learning
mathematics more fun for children in
Grades four through seven. It is at this
age that a lot of students become
turned off to math and science
subjects, and this spells trouble for the
future, when people skilled in science
and engineering are going to be in
even greater demand.
The research team involved in the
educational project draws from
various disciplines, including Computer Science, and the Faculty of
Education. Professional video game
designers are working on the project
as well. According to Klawe, the
problem is a complex one. "It's really
not known what kids learn when they play
video games and what types of things they
can learn in that environment."
On the plus side, she says discrete math is
easy to communicate and visualize, and
doing so will make the concepts more
concrete for school children. "We're going
to make it come to life for them," said
Another unrelated, but major focus of
Klawe's research is on constructing certain
kinds of data structures to maximize
efficiency in accessing alphabetical data.
In a given alphabetical list, there are
usually items system users want to access
much more often than others. "I want to
optimize that structure," said Klawe.
The technical term for the subject is
"optimal alphabetic binary search trees."
Klawe said it is one of the long-standing
problems in Computer Science, but that
she, with the help of a graduate student,
has made significant progress. "We now
have optimal time solutions for some
classes of problems that arise in practice,"
said Klawe.
She and members of her department are
looking forward to tackling more research
problems once the new CICSR-Computer
Science Building opens this summer. The
building will help address the severe
Computer Science Department Head Maria
Klawe is working on an interdisciplinary
project to develop educational videos that
make it easier for young students to grasp
mathematical concepts.
shortage of space her Department now
suffers from due to phenomenal growth in
recent years. Both the Computer Science
faculty and graduate student numbers have
doubled over the past five years, and
undergraduate enrollment is up some 40
per cent from last year alone.
"The new building will give us space to do
many projects we've wanted to do, but had
to put off," said Klawe. She adds that it
will allow room to bring in visiting
researchers. In the past, the Department
has had to turn down many requests from
computer scientists wanting to visit as
researchers at UBC.
Another advantage of the new building will
be the increased opportunities for interdisciplinary research with other CICSR
members in Mechanical Engineering and
Electrical Engineering. "There are so many
research areas where we're at the boundary
between Engineering and Computer
Science," said Klawe. Yet another advantage is that graduate students often want to
take courses in both areas, and will find it
easier when the departments are no longer
half a mile apart. Generally, the arrangement will allow for increased and practical
interdisciplinary research, a concept Klawe
supports wholeheartedly. ■ ASI promotes collaboration
ASI executive director Brent Sauder is working to give industry
easier access to university research.
Brent Sauder, who joined the B.C.
Advanced Systems Institute in October
1992, is working to give industry easier
access to university research. Sauder left
MacMillan Bloedel Research after 12 years
to become ASI's new executive director.
He plans to maintain and strengthen ASI's
ties with CICSR. ASI provides fellowships
for 11 CICSR members. ASI and CICSR
share several common goals, including the
desire for more linkages between university and industry. ASI fellow are required
to work directly with companies as part of
their applied research.
"There is a fair amount of ignorance,
especially among smaller companies,
about the strong research capabilities that
exist at universities," said Sauder. "These
companies could use universities to meet
their R&D needs."
He adds that at the initial stage of a
university research project, the choice of
platform or development language may be
quite arbitrary. He would like to initiate
dialogue between industry and university
researchers at the start of the project so
that the result are in a form industry can
really use.
Sauder is working hard to get what he calls
more "technology pull" from industry to
guide applied research. He would like
industry to come to ASI with descriptions
of problems advanced systems could solve,
and then match these problems with
expertise within ASI. In most cases, both
sides of the equation benefit: industry gets
its problem solved by experts at a reasonable cost, and researchers are provided
with funding for their work, which they get
to see applied.
There are obstacles to increased university-
industry collaboration, however. Technology companies often don't have the time
B.C. Advanced Systems Institute executive
director Brent Sauder (centre) with James
Varah (right) and Kelly Booth of CICSR.
According to Sauder, the fellowships give
researchers relief from teaching, which
provides the time needed to set up relationships with industry. Fellows also find that
with a ASI fellowship in hand, it is much
easier to attract outside funding.
In the future (provided the B.C. government continues to support ASI), Sauder
plans to introduce new programs aimed at
stepping up university-industry collaboration. These include giving ASI affiliate
companies the option to apply ASI funds to
a specific university research project they
need done.
Sauder would also like to expand the
visiting fellows concept. ASI now brings
experts from around the world to lecture or
work for a limited period of time at
universities to share their expertise. Sauder
would like to see a similar program for
fellows to spend time with companies
transferring their knowledge.
"ASI should not be viewed as a granting
agency, but as a solutions provider," said
Sauder. "We have really strong ties to the
universities. Now we have to do the same
with industry. If we want a high-tech
industry here in B.C., we need to become
team players." ■
"We have really strong ties to the universities. Now we have to do
the same with industry. If we want a high-tech industry here in
B.C., we need to become team players."
required to find the right people at the
universities and to work with them. As
well, company representatives don't
always understand the goals and priorities
of university researchers. They may be
concerned about ownership of the intellectual property resulting from a research
project they sponsor. However, says
Sauder, if the technology is crucial, there
are ways to work these concerns out.
Sauder has found fewer concerns from the
university side of the equation. "ASI
fellows have a real drive to get out there
and do something applied and relevant.
The ASI faculty is really interested in
solving industry problems." In fact, many
ASI fellows from CICSR work with not
just one, but several companies on various
research projects.
CICSR's ASI Fellows
Hussein Alnuweiri*
Kelly Booth
Clarence de Silva
Guy Dumont
Dave Forsey
Alain Fournier
Andre Ivanov
Nicolas Jaeger
Jeffrey Joyce
Samir Kallel*
Takis Mathiopoulos
Matthew Palmer*
Tim Salcudean
Carl Seger
* ASI New Faculty Awards Vision of the Future
CICSR researcher David Lowe is developing vision system to be
incorporated into industrial robots of the future
People tend to take the ability to see for
granted. It's a sense we rely on so much,
we don't really think about how it works.
What seems deceptively simple, however,
is extremely complex. The human vision
system is actually an integration of about
50 separate systems, including stereo
vision matching, edge detection, motion
perception, shape recognition, colour
analysis, contrast adjustment and more.
Vision takes up a full 25 per cent of our
brain power.
What David Lowe of the Department of
Computer Science is working on is
replicating a simple version of the human
vision system on a computer. Of the many
vision systems that make up our ability to
see, Lowe is concentrating on giving
computers the ability to recognize objects.
It's an important application that will make
industrial robots far more productive.
Currently, the major cost of installing a
robot on a production line is the cost of
ensuring that the parts the robot works
with are in precisely the right places.
Without the ability to see, a robot can do
its work only if it can move to exactly the
same place every time to perform its task.
If the part it's working on gets shifted at
all, the robot can't adapt, and the whole
system stops working.
Vision systems require tremendous
amounts of computing power. It takes
some $50,000 worth of computer hardware
to make the vision system in the UBC lab
work. But the costs are coming down to
the point where vision is viable for more
widespread applications. Lowe has worked
with a small company, Kinetic Sciences
Inc. (KSI), for two years developing a
vision system for the robotic system KSI is
developing for the space station.
The space station will have several robotic
systems, including the Canadarm, to help
build and maintain the structure in space.
Robots will also perform routine tasks to
free up valuable astronaut time. According
to Lowe, there are a lot of tasks the robotic
systems in space should be able to do on
their own, but a good vision system is a
prerequisite to doing a lot of automation.
Part of Lowe's work in transferring his
knowledge to industry is to make vision
systems more reliable and affordable. He
is also working on ways to improve
computer recognition of objects, no matter
what their orientation is in space. Think
about how different a pencil looks if you
David Lowe of the Department of Computer
Science is working on development of computer
vision systems that are affordable and reliable.
see it with the point facing straight at you,
and then see it turned sideways so the full
length is visible. How does the computer
know both objects are the same?
Lowe is developing algorithms to enable
computers to learn from their experience.
These algorithms are based on statistical
techniques similar to the way the human
brain operates. The goal is to develop a
vision system that enables the computer to
recognize objects faster the more often they
see them.
Today's industrial vision systems are fairly
simple. They can recognize industrial parts
and rigid objects. But the potential is
enormous. Lowe believes that in the future,
virtually all mechanical systems will have
vision. Futuristic vision products now in
the works include vacuum cleaners that run
themselves, computer terminals that follow
your eye movements negating the need for
a mouse, and cars that automatically avoid
collisions. Many of these products are
already in prototype form.
But computer vision research has to
advance considerably to make some of
these products a reality. Lowe expects his
work to be applied first to automate
repetitive industrial tasks such as construction work, and loading and unloading. He
works with researchers in other disciplines
to integrate vision into robotics and
computer intelligence.
"It's a good CICSR project," said Lowe.
"The future of this type of research is to
integrate systems." He is working with
other CICSR researchers towards the goal
of developing a machine incorporating
vision, mechanical engineering and
computer reasoning into one complete
system. Vision by itself isn't all that useful,
says Lowe.
If Lowe could change one thing about his
field, it would be the general perception of
it. He says the biggest problem in the field
of computer vision is the fact that people
don't understand the true complexity of the
problems he and other researchers are
trying to solve. People are equipped with
incredibly sophisticated vision systems that
work so well, we take them for granted.
After years of working to get computers to
emulate our vision system, Lowe looks at
the world with a much better appreciation
of what he can see. ■ 8
UBC appoints top engineer as
Associate VP Research
■ Martha Salcudean, head of the
Mechanical Engineering Department,
has been appointed Associate Vice
President, Research. The appointment is
effective November 1, 1993.
Salcudean will help promote and
develop a broad spectrum of basic,
applied and targeted research activities.
She will also liaise with government,
industry and other agencies.
She is a noted researcher who has won
acclaim for her computer simulations of
fluid flow in industrial processes. Her
I research funding is among the highest of
any engineering researcher in Canada. ■
Start Planning Now For
Science & Technology Week
■ Planning is already under way for
Science and Technology Week '93,
slated for October 16-24. The theme of
this year's event will be: "Discover the
Scientist in You."
S&T Week is designed to enhance
public awareness of the role science and
technology plays in our lives and the
importance of these industries and skills
to our future. To find out how you can
particpate, contact Ingrid Smith, Science
and Technology Week Coordinator,
Ministry of Advanced Education,
Training and Technology, Science and
Technology Div., tel: (604) 387-1628. ■
New CICSR lecture series
■ CICSR will be hosting a new lecture
series starting this fall featuring lectures
by CICSR faculty members. The CICSR
Faculty Forum is a great opportunity to
hear about world-class research being
performed right here at UBC. ■
Distinguished Lecture Series 1993-94
Parallel and Distributed Systems
Six academic and industrial leaders
address the future of computing: Parallel
systems, ubiquitous computing, and more.
September 9,1993
Specifying Concurrent Systems
Leslie Lamport
Digital Equipment Corporation, Systems
Research Centre
October 14,1993
How to Design a Parallel Computer
David May
Inmos Co., UK
November 18,1993
Ubiqitous Computing: Origins, Current
Research, and the Future
Mark Weiser
Xerox Palo Alto Research Centre
February 10,1994
New Abstraction for Scalable, Portable
Parallel Programming
Lawrence Snyder
University of Washington
March 10,1994
Shape Recognition in Neural Networks
Geoffrey Hinton
University of Toronto
April 14,1994
Omega Networks, Combining and the
Ultra III Prototype
Allan Gottlieb
New York University
CICSR is hosting its sixth annual Distinguished Lecture Series, bringing in
academic and industrial leaders in the
forefront of their respective fields.
This year, DLS speakers will be discussing
computing systems of the future, including
parallel computing, neural networks and
ubiquitous computing. DLS speakers will
provide a picture of the latest computing
techniques and provide clues to where this
fast-moving field is heading.
Join us for a Glimpse of the Future of
Computing Systems
Lectures are from 4:00 pm to 5:30 pm, in
the new CICSR/CS Building's large
seminar room, 2366 Main Mall, UBC.
Lectures are complimentary.
The UBC Centre for Integrated Computer
Systems Research (CICSR) is an interdepartmental research organization made up of
computer-related research faculty members in
the Departments of Computer Science, Electrical
Engineering and Mechanical Engineering.
Currently there are more than 60 CICSR
researchers which direct over 200 graduate
students and collaborate with dozens of
industrial firms in areas such as robotics,
artificial intelligence, communications, VLSI
design and industrial automation.
CICSR FOCUS, is published twice a year.
EDITOR:  Leslie Ellis
DESIGN:  Rob Bishop
Office: 2053 - 2324 Main Mall,
Vancouver, B.C. V6T 1Z4
Tel: (604) 822-6894, fax: (604) 822-9013
Contact: Gale Ross


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items