Multi-agent systems are winning more and more battles in the artificial intelligence field. A feasible and successful approach proved to relay on small and simple agents inspired from biological models. This approach emphasises the significance of the environment in the agent system, the attention that must be paid to the dynamics that emerges from interactions and the importance of growing rather then building agents. Finding out the form and parameters that influence the system behaviour is non-trivial for humans, even more difficult than knowledge engineering. The paper proposes and investigates the methods that are enabling user-driven solutions in dealing with this challenge. The methods are based on two essential functional requirements: a) at macro level, the need to monitor and represent in an intuitive way the system behaviour and b) at micro level, the need to control and track the system state space. The approach proved to be workable on usual configurations and effective in dealing with combinatorial explosion.
Stigmergic Coordination, Multi-Agent Systems, Synergy, Symbolic Paradigm, Operations Research, Travelling Salesperson Problem.
All articles published by Studies in Informatics and Control (SIC) are published in accordance with the terms and conditions of the Creative Commons Attribution License (CC BY-NC-SA 4.0).