|
|
-
Double time
Two jobs, two NSF grants. Everything seems to be in twos for Robyn
Lutz.
-
Robyn Lutz is used to doubling up on things.
She is not only employed as an associate professor of computer science
at Iowa State, but also as a senior engineer for the Jet Propulsion Laboratory
in Pasadena, Calif. This facility, administered by Caltech for NASA, focuses
on unmanned space exploration.
So it should come as no surprise that Lutz is doubling up on grants from
the National Science Foundation (NSF).
But what may come as a surprise is the time frame for her grants.
Lutz received notification from the NSF just a week apart last July that
she had not only been awarded her first NSF grant, but a second one as
well. The two grants started on the same date (Sept. 1, 2002) and expire
just a month apart (August-September 2005).
"Both (NSF grants) are related to software engineering and the safety
analysis of software systems," Lutz said.
Safety analysis and software is something that Lutz has been working on
for the past 19 years at the Jet Propulsion Laboratory and now at Iowa
State. Because technological advances and the consumer marketplace are
producing increasing numbers of safety-critical applications, Lutz says
it’s important to realize that such advances can be both positive
and negative.
"Software can contribute to a system's safety or can compromise it
by putting the system into a dangerous state," she says. "Software
engineering of a safety-critical system requires a clear understanding
of the software’s role in, and interactions with, the system."
Lutz says examples of safety-critical software include the software for
implantable medical devices, for smart vehicles, and for industrial robots.
For Lutz, however, her greatest interest lies in spacecraft that explore
deep space and voyage to other planets.
"Historically, many failures of safety-critical software have been
due to an inadequate understanding of the software requirements by the
developers," she said. "The specification of what the software
had to do was incomplete or inconsistent in some way that prevented safe,
correct software from being developed."
Lutz's pair of NSF grants will look into these concerns.
In her three-year, $279,000 grant on "Safety Analysis for Critical
Product Lines," Lutz will address the question of how safety analysis
can become a reusable asset of a product line by developing a framework
and a suite of techniques for the safety analysis of critical product
lines.
Lutz cites an example of a product line in the airline industry.
"Each airline wants a slightly different airplane display panel,"
she said. "We have to be aware that the software doesn't insert hazards
into the system because of these differences."
The second NSF grant (a three-year, $202,000 grant on "Natural Language
in the Development of High-Confidence Software") looks at inadequate
communication of domain knowledge in natural language (such as English
textual descriptions) as a major source of requirements defects in high-confidence
software.
"Such defects can threaten lives, property and the dependability
of critical infrastructures," she said. "This research develops
innovative, multi-disciplinary techniques designed expressly to identify
and cope with the properties of natural language that lead to these problems."
Such a defect is thought to be the cause of the failed Mars Polar Lander
mission. The software thought that the spacecraft had landed on Mars and
turned off the spacecraft's engines. Instead, the Mars Polar Lander was
still above the surface of Mars, crashed and was lost.
"My interests in this area are within the space program, but these
techniques are broadly applicable to other industries," Lutz said.
Lutz's contributions in this area have netted her not only two NSF grants,
but a recent award from the NASA Office of Safety and Mission Assurance.
She received an award for "outstanding contributions to software
assurance research."
Around LAS
November 4-17, 2002
|
|
