Intelligent Interactive Technology
RoboCup: Complex Systems
Complex systems characterise a large number of real-world problems that exist in almost all areas of our life, including management of emergencies, networked digital appliances and services, business and commerce, natural phenomena and many others. Usually, a complex system is characterised as follows:
- a medium-sized number of agents;
- agents are generally both intelligent and adaptive (they make decisions in accordance with various rules and are ready to modify their rules of action on the basis of new information);
- there are no centralised controllers (no dictators);
- agents make their decisions and update their action rules on the basis of local, rather than global, information (no single agent has access to what everyone else is doing - autonomy).
Our RoboCup research is motivated by the shift in the AI research community towards a
study of complex systems and real-world problems.
Cyberoos is a team of
synthetic (software) agents designed by our group to compete in the RoboCup Simulation
League - an artificial multi-agent domain. Cyberoos2001 is the fourth
generation developed to
provide a test-bed for experiments with emergent tactical teamwork.
To learn more about our RoboCup effort and challenges we face, please contact Mikhail Prokopenko. In summary, our research applies multi-agent technology in the following areas.
Teamwork Simulation and Real-time Automated Reasoning
Intelligent Networked Digital Appliances
Search and Rescue Planning
Other Applications
RoboCup Rescue is a new initiative, which addresses search and rescue operation planning and execution for large-scale disasters such as earthquakes or forest fires.
To toplast updated
July 11, 2005 05:40 PM
Mikhail.Prokopenko@csiro.au