iClips: an interaction system for controlling a robotic FACE.

Expert systems have reached large popularity since they were born in the 1980s. They are systems which encode human expertise in limited domains by representing human knowledge using if-then rules. Since one of the goals of Artificial Intelligence is to develop systems which exhibit "intelligent" human-like behaviour, expert systems fully find their way in this field. iClips, the system developed in this work, is an interaction tool for controlling a robotic FACE. It takes its name from the existing CLIPS, an expert system developed from the NASA Space Centre that provides a complete environment for the construction of rule-based expert systems. iClips aims to extend the existing robotic project by implementing the cognitive part of the android through the introduction of a modular architecture which allows the user to coordinate, control and debug it. Starting from a monolithic library, one of the goals iClips manages to reach is to build a deliberative/reactive two-levels architecture for coordinating the system and having a symbolic representation of our targets and purposes. By using symbolic abstraction iClips turns out to be a declarative tool and, due to this, it can be easily used by hybrid subjects, that means subjects who don't have very specific computer skills: while building a new module requires a programmer, working with iClips requires people who are just able to write simple logic rules