An Adaptive Genetic-Based Architecture for the On-line Co-ordination of Fuzzy Embedded Agents with Multiple Objectives and Constraints

Abstract

This paper presents a novel embedded agent architecture that aims to co-ordinate a system of interacting embedded agents in real-world intelligent environments using a unique on-line multi-objective and multi-constraint genetic algorithm. The embedded agents can be complex ones such as mobile robots that would operate hierarchical fuzzy logic controllers or simple ones such as desk lamps that would bear threshold functions instead. The architecture would enable the agents to learn the users' desires and act based on them in real-time without having to repeatedly configure the system. The system can handle unreliable sensors and actuators as well as compensating for agents that break down and adapting on-line to sudden changes. The architecture allows for the organisation of agents to be dynamic since it accommodates for agents migrating in and out of the system. Multifarious experiments were performed on implementations of the aforementioned architecture where the system was tested in different scenarios of varying circumstances