| Modern information system environments are becoming large, open, and heterogeneous. The software agent technology plays a key role to address specific problems such as Resource Management, Knowledge Management, Security, Interoperability of multiple systems, etc. There are certain limitations in the agent paradigm. One of the major limitations is the lack of proper semantics for distributed agent communication. Because of the dynamic nature of multi-agent environment, for agents to be able to understand the context in which they are operating, there is a need for proper semantics. This thesis proposes a set of semantics, called Knowledge Based Communication Language (KBCL) for communication between agents within multi-agent applications. Key issues in designing, programming and establishing communication between agents are identified. A meaningful communication between multi-agents is established through a common Ontology. A framework for agent based communication is developed and it is demonstrated for a practical application using the Knowledge Based Communication Language for a Water Reservoir Management System. |
