Prometheus: A Methodology for Developing Intelligent Agents

摘要

We have briefly described one step of the Prometheus methodology. The methodology has been in use for several years and has been taught in industry workshops (most recently at the Australian AI conference, 2001) and to students at RMIT University. The feedback we have received indicates that it provides substantial guidance for the process of developing the design and for communicating the design within a work group. One of the advantages of this methodology is the number of places where automated tools can be used for consistency checking across the various artifacts of the design process. For example, the input and output events for an agent must be the same on the system overview diagram and on the agent overview diagram. Agent Oriented Software Pty. Ltd. has constructed a support tool for the methodology that allows design diagrams to be drawn and generates corresponding skeleton code in JACK. Future work includes clearer integration of goals, extension to social agent concepts (e.g. teams, roles), and investigating the use of design artifacts in debugging. We are also in the process of writing a book on the design of intelligent agent systems. This will be a practical "how-to design agent systems" aimed at industrial practitioners and undergraduate students.