Beruflich Dokumente
Kultur Dokumente
(1991) (© AAAI)
Workshops
approaches to design problems and small-waterplane-area twin-hull approaches and open research issues.
critical research issues in the field; (SWATH) ships, (2) current research The problem was to design a SWATH
current AI methods for reasoning in computer-aided acquisition and ship whose primary purpose is to act
with physics knowledge; research analysis of hydrodynamic knowledge as a helicopter platform for landings
progress in computational methods related to ship-wave motions, and (3) and take-offs at high seas; given char-
for the exploration and analysis of computer system requirements for acteristics of speed, range, and so on,
physical systems; and experimental ship design. A report on the HOSDES had to be satisfied. The problem was
AI design systems in several domains system for ship hydrodynamic design, presented as an example of realistic
—mechanical engineering, digital cir- which is currently in the advanced design situations that are significant
cuit design, and the design of special- stages of development at MARIN, for Navy planning and that can ben-
purpose computer architectures. A gave a good sense of the best com- efit from AI approaches. Also, it illus-
progress report on the development puter environment for ship design trated some of the challenges that
of the Engineous design system at GE that can be attained with current problems in this area offer to AI
(which combines AI approaches and technology. researchers, in particular, developing
computational fluid dynamics [CFD] The last one-third of the program a concurrent design with both hydro-
analysis tools) and its use in aircraft was devoted to brainstorming ses- dynamic and structural considerations
engine design provided an up-to-date sions that consisted of technical dis- and acquiring and using new hydro-
view of the application of AI technol- cussions about the nature of ship dynamic knowledge about wave-ship
ogy to problems that have many ele- design problems, the requirements motion, knowledge that is needed to
ments in common with ship for advanced computer environments provide insight into behavioral char-
hydrodynamic design. in this area, assessments of relevant acteristics of proposed designs and to
Presentations in the general area of AI research issues and possible AI guide design decisions.
computational approaches to ship approaches to hydrodynamic ship In a luncheon talk, Bob Simpson
design and related work in computa- design, and ideas about futur e presented a DARPA perspective on
tional hydrodynamics covered (1) research. research planning, with an emphasis
detailed reconstructions of decisions To focus the discussion, a specific on advances in AI and the related
and experience in the computer- ship design problem was formulated area of information science and on
aided design of racing yachts (in par- by Nils Salvesen and Marty Fritts issues of industrial productivity and
ticular, the Stars and Stripes, whose from SAIC. The problem provided a competitiveness. Dinner talks were
victory in the 1987 America’s Cup concrete basis for clarifying concepts given by Alexander Pond, executive
race owes much to the intensive use and methods across disciplines and vice president of Rutgers, and Raj
of computers in its design) and of presenting specific ideas about AI Reddy, from CMU, then president of
SPRING 1990 15
Workshops
SPRING 1990 17