Self-Controlled Production: Forecasting with Smart Sensor Technologies

Wolfgang Maaß, Shahd Zahran

In: 23rd Workshop on Information Technology and Systems (WITS), Milan, Italy. Workshop on Information Technology and Systems (WITS) Springer 2013.


Sensor-based architectures are currently simple data collectors that feedforward to central data processing units. With huge amounts of sensors data networks are threatened by congestion of low-quality data and noise. Therefore we present a Smart Sensor Architecture that enables semantic annotation of sensor data near the local environment of a sensor and processing of data by sophisticated algorithms. We show technologies by which the Smart Sensor Architecture is realized and discuss how the behavior of smart sensors can be changed at run-time. Our approach was deployed in a plant irrigation scenario. Initial results of the evaluation are presented.

Weitere Links

happy-plant-demo-v20.pdf (pdf, 4 MB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence