A Multi-Agent-Based Intelligent Sensor and Actuator Network Design for Smart House and Home Automation

The smart-house technology aims to increase home automation and security with reduced energy consumption. A smart house consists of various intelligent sensors and actuators operating on different platforms with conflicting objectives. This paper proposes a multi-agent system (MAS) design framework to achieve smart house automation. The novelties of this work include the developments of (1) belief, desire and intention (BDI) agent behavior models; (2) a regulation policy-based multi-agent collaboration mechanism; and (3) a set of metrics for MAS performance evaluation. Simulations of case studies are performed using the Java Agent Development Environment (JADE) to demonstrate the advantages of the proposed method.