Wireless Sensor Networks are an integral part of the field of Ambient Intelligence, in that they can form a cheap, easy-to-deploy and infrastructure-less system for sensing and actuating. However, the distributed nature of the system and the limitations of individual sensor nodes raise the challenge of how to organize such a system and how to build a meaningful application for it.

In this presentation it will be argued that AI-inspired methods (machine learning, multi-agent systems, bio-inspired algorithms) are the way to go when tackling these challenges. Some example implementations on an existing WSN platform will be presented, as well as some currently open research issues.