Applications of intelligent agents
Agent technology
From AOSE methodology to agent implementation
Proceedings of the first international joint conference on Autonomous agents and multiagent systems: part 1
Introduction to Multiagent Systems
Introduction to Multiagent Systems
Machine learning and inductive logic programming for multi-agent systems
Mutli-agents systems and applications
Artificial Intelligence: A Modern Approach
Artificial Intelligence: A Modern Approach
Developing Intelligent Agent Systems: A Practical Guide
Developing Intelligent Agent Systems: A Practical Guide
Proceedings of the fourth international joint conference on Autonomous agents and multiagent systems
Prometheus: a methodology for developing intelligent agents
AOSE'02 Proceedings of the 3rd international conference on Agent-oriented software engineering III
A multi-agent architectural solution for coherent distributed reconfigurations of function blocks
Journal of Intelligent Manufacturing
An intelligent supervision system for open loop controlled processes
Journal of Intelligent Manufacturing
| Hi-index | 0.00 |
Mass production, such as white goods manufacturing, is traditionally bound to hierarchical factory-floor procedures and accepts only gradual changes in technology and product architecture. This paper introduces an idea on how to upgrade from classical to network-connected reconfigurable devices. A generic multi-agent architecture was created, derived from belief-desire-and-intention (BDI) agents. It covers all types of white goods in the form of rational home assistant, and enables reconfiguration of agent-based household appliances during the design, production and implementation phases. The introduced concept involves a multi-agent architecture which utilizes distributed processing power at different levels: higher-level agents run on more powerful devices than embedded appliance's controllers, personal assistant (PDAs), or Windows or Linux based personal computers (PCs). PDAs can run a single agent, for example a GUI agent, whereas the embedded controllers execute lower-level device (embedded) agents. In this way, all the appliance's basic functionality, such as its hardware units (e.g., electrical motors, valves, heaters, etc.), are initially simulated by auxiliary agents running together with higher-level agents on a PC or PDA. Using this simulator in the design phase, all vital functions and capabilities of the agent-based appliance under development are thoroughly tested first. Afterwards, the agents that simulate the device's hardware units and environment are simply replaced by the communication to the corresponding device units. In such way, any new functionality or device's behaviour can be upgraded any time just by adapting the core of the multi-agent architecture on the PC and individual agents on the PDA or the embedded agents in appliances. A thorough design and implementation cycle of the proposed solution using two freeware development tools is also described, i.e., the Prometheus agent design methodology and the agent simulation/execution environment called Jadex. The approach is exemplified by building a simulator of an agent-based household appliance, namely a Multi-agent Washing Assistant as a special instance of rational home assistant.