Augmenting BDI with relevance Supporting agent-based, pervasive applications

The potential of pervasive mobile systems lies in the integration of applications and services available on mobile devices. These services must be built on the premises of contextawareness, such that the application takes advantage of environmental information to improve its performance. The challenge is to process more environmental information with the limited computational resources available on mobile devices. Intelligent Agents have been proposed as a suitable architecture for these tasks. Such agents are designed around a deliberation cycle. We show that one of the problems agents encounter in highly dynamic environments is that this cycle misses opportunities to react to the changing environment. For this reason we propose an extra module that augments agent architectures with a “context-observer”. Based on the structures used in the deliberation cycle, this module evaluates the relevance of received data. It does this parallel to the normal cycle. We illustrate how such an approach could improve the context sensitivity of the agent’s behaviour and thus the usefulness of the application.