Composing Cognitive Agents from Behavioural Models in PRESTO

The PRESTO project applies agent technologies to serious gaming. PRESTO has developed an AI infrastructure and an agent framework called DICE for the creation of gameindependent, modular Non-Player Characters (NPC) behaviours based on a BDI (Belief-Desire-Intention) approach enriched with cognitive extensions for human simulation. Behavioural models can be combined via end-user development tools to form the behavioural profiles of NPCs in a game. DICE provides the coordination between body-controlling behavioural models (for navigation as well as posture, facial expressions, actioning) and decision-making models representing e.g. the standard operating procedures of professional roles, the cognitive appraisal of events and perceptions, the modality of reaction to unplanned events occurring during a game. Behavioural models are largely if not completely independent of the specific scenario or even game engine thanks to abstractions of both the environment and the internal state of the NPC provided by means of ontologies. PRESTO is producing a set of behavioural models targeted at its pilot project’s needs or expected to be of common use, including navigation in the virtual environment sensitive to the cognitive state of the NPC. This paper gives a brief overview of PRESTO, DICE and of its ontologies.