Self planning in critical multi-agent systems

Abstract

Novel architectures like Cloud and Internet of Things (IoT) make available several resources like computing nodes, environmental sensors etc. that enable the introduction of more and more intelligent systems able to face complex situations. In particular management of critical and dangerous situations may take advantage of those systems whose complexity is growing up faster and faster. In this scenario, it is difficult to orchestrate different autonomous systems in order to face with new, previously unmanaged emergencies. In this work we present a modeling methodology and a planning techniques based on a multi-agent model. Agents describe capabilities of each available IoT element in an area where a critical situation has occurred; the planning methodology exploit both classical and a new counter-example based approaches to build a coordination plan of resources in order to achieve given goals like traffic management or people flight during a terrorist attack.