In a world of multi-stakeholder information and assets provision on top of millions of real-time interacting and communicating things there is a demand of enhancing sustainability of smart city applications by allowing IoT based systems to reach their full potential. This talk digs into the development of tools and functionalities to evolve and act in a more autonomous way, becoming more reliable and smarter.

Things will be able to learn based on others experiences, while situational knowledge acquisition and analysis will make things aware of conditions and events potentially affecting their behaviour. Adaptive selection approaches will manage the uncertainty and volatility introduced due to real-world dynamics.

COSMOS Project is working in the integration of decentralized management mechanisms in IoT based systems allowing applications to exploit millions of things. Socially-enriched coordination will consider the role and participation scheme of things in and across networks. Management decisions and runtime adaptability will be based on things security, trust, administrative, location, relationships, information, and contextual properties. Thus the resulting platform overcomes current drawbacks and targets novel objectives that goes beyond state of the art solutions.

The validation of such functionalities and technologies will be done through the implementation of two use cases in real scenarios. The first one is going to be deployed in Madrid and the objective is to introduce a system that helps in the handling of passengers with special needs. The second scenario covers Candem district and its aim is to improve the efficiency of heating system. The tools in both cases are IoT devices with learning capabilities. The combination of IoT features and cloud based interactions will result in a win-win situation for citizens and service providers in cities.