# Australian Artificial Intelligence Institute

The Australian Artificial Intelligence Institute (AAII) was a government-funded research laboratory in Melbourne, active from 1988 to 1999, focusing on intelligent software agents and producing systems like PRS, dMARS, and SWARMM.

The Australian Artificial Intelligence Institute (AAII) was a government-funded research and development laboratory based in Melbourne, Victoria, that operated from 1988 until 1999. Its primary focus was investigating and commercialising artificial intelligence, with a particular emphasis on intelligent software agents. The institute produced several notable software systems, including a procedural reasoning system (PRS), a distributed multi-agent reasoning system (dMARS), and a Smart Whole AiR Mission Model (SWARMM).

## History

The AAII was established in 1988 as an initiative of the Hawke government, with backing from the Computer Power Group, SRI International, and the Victorian State Government. The founding director was Michael Georgeff, who came from SRI International and brought experience with the PRS and a vision for intelligent agents. Initially located in the Melbourne suburb of Carlton, the institute later moved to larger premises in the city centre, occupying two floors of an office building on the corner of Latrobe and Russell Streets. At its peak, the AAII employed more than 40 staff.

In the late 1990s, the institute spun out Agentis International (later Agentis Business Solutions) to commercialise its developed technology. Additionally, several former staff members formed Agent Oriented Software (AOS) to pursue agent technology, developing JACK Intelligent Agents. The AAII closed in 1999, and the remaining staff and intellectual property were transferred to Agentis International.

## Projects

The AAII engaged in a range of software and commercial projects. The procedural reasoning system (PRS) was developed and applied in collaboration with SRI International. The distributed multi-agent reasoning system (dMARS) was an agent-oriented development and implementation environment for building complex, distributed, time-critical systems, developed as a C++ extension to PRS. The Smart Whole AiR Mission Model (SWARMM) was an agent-oriented simulation system developed with the Air Operations Division (AOD) of the Defence Science and Technology Organisation (DSTO).

The Optimal Aircraft Sequencing using Intelligent Scheduling (OASIS) system was an air traffic management system written in PRS that estimated aircraft arrival times, determined optimal landing sequences, and alerted operators to required actions. It aimed to reduce congestion and maximise runway use. A prototype for Sydney Airport, called HORIZON, was developed using dMARS. The Single Point of Contact (SPOC) system, built for Optus, assisted customer service representatives in meeting the objective of resolving 98% of customer enquiries with a single point of contact, using a multilayer architecture based on dMARS.

## Technical notes

Over its existence, the AAII released more than 75 public technical notes. Notable examples include Anand S. Rao and Michael P. Georgeff's "Asymmetry Thesis and Side-Effect Problems in Linear-Time and Branching-Time Intention Logics" (AAII Tech Note 13, 1991), "Modeling Rational Agents within a BDI-Architecture" (Tech Note 14, 1991), and "Intelligent real-time network management" (Tech Note 15, 1991). Michael P. Georgeff authored "Situated Reasoning and Rational Behaviour" (AAII Tech Report 21, 1991). Rao and Georgeff also published "BDI Agents: From Theory to Practice" (AAII Tech Report 56, 1995), which became a widely cited work in the field of [belief-desire-intention](https://www.wikiprompt.org/wiki/belief-desire-intention) software models.

## Legacy

The AAII contributed significantly to the development of intelligent agents and the [belief-desire-intention](https://www.wikiprompt.org/wiki/belief-desire-intention) (BDI) architecture, influencing subsequent research and commercial applications. Its work on procedural reasoning systems and [multi-agent system](https://www.wikiprompt.org/wiki/multi-agent-system)s laid groundwork for later technologies. The institute's closure in 1999 marked the end of an era, but its intellectual property and personnel continued through spin-off companies, ensuring the continuation of its research directions.

## See also

- Distributed multi-agent reasoning system
- [Belief-desire-intention](https://www.wikiprompt.org/wiki/belief-desire-intention)
- Procedural reasoning system

## External links

- [Wikipedia: Australian Artificial Intelligence Institute](https://en.wikipedia.org/wiki/Australian_Artificial_Intelligence_Institute)

---
Source: https://www.wikiprompt.org/wiki/australian-artificial-intelligence-institute
License: CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0/)
Last updated: 2026-09-14T04:20:04.130444+00:00