Extending the bdi-asdp methodology for real-time

Abstract

Multi-agent systems are an emerging research area which is experiencing a fast growth. In the last years, many theories, architectures, languages and platforms for the development of agent based systems have been developed. On the other hand, real-time systems represent an important challenge from the perspective of multi-agent systems, considering the increasing need to count on software which is able to respond to certain situations in a timely partition. Nevertheless, in spite of its increasing interest, an important difficulty when applying these technologies to the resolution of a concrete problem is in the agent-based software development process. Many efforts have been made to extend the capacities of standard software modelling to integrate multi-agent and real-time systems, and different methodological proposals exist, but their practical application is not always obvious from the definition of the methodology. This work proposes an extension of the BDI-ASDP methodology for the inclusion of timing constraints. We have applied this methodology to model some of the agents which participate in a virtual baseball game.