Intelligent Agents in the Australian Bureau of Meteorology
Sandy Dance and Malcolm Gorman
9-30am - 11am,
Room 2.16 (level 1 staff tea room)
SEECS Building,
221 Bouverie St.
The University of Melbourne
Carlton
Abstract:
The Australian Bureau of Meteorology is the national weather service
for Australia. It has a strong need for complex and evolving systems
for managing its weather forecasting, monitoring and alerts and it is
currently in the process of updating and reviewing software and
processes to this end. The rapidly growing area of intelligent agent
technology is an excellent candidate for the kind of distributed,
complex and open system that is required. However, in order for
intelligent agents to be successfully used in this kind of environment,
there are some fundamental questions that must be addressed involving
issues such as how agents will find and use newly added services and
how services will communicate with each other, given that they are
developed independently.
The bureau environment is an excellent test bed for these fundamental
research questions, as it is a highly computerised
information processing organization, with a mixture of existing systems
and humans. An intelligent agent operating in this system would be
able to act autonomously, making the best decisions and finding the
best possible way of achieving goals in the given environment.
The key objective of the project is to develop the required mechanisms
to allow agents to locate and make use of data and services in a
distributed open system. Specific technical objectives are:
- To develop system infrastructure and architecture that facilitates
agents (and other software) working together.
- To develop mechanisms that support dynamic and evolving systems of
agents where software agents may be added - or removed - at any time.
- To identify appropriate languages for description of data to allow
sharing and understanding of data by agents with differing but related
ontologies.
In this talk we describe the project which is intended to fulfill these
objectives.
Biosketch:
Sandy Dance received his BSc from Monash University in 1969 and his
MSc in mathematics from the same institution in 1972. For a number of
years he worked in the commercial computer industry before returning
to study for his PhD at the University of Melbourne, which he obtained
in 1995. After two years as a post-doctoral fellow, he moved to
the Australian Bureau of Meteorology Research Centre, where he is
working with weather radar. His interests lie in the areas of high
level interpretation of images, artificial intelligence and agent
networks