constitute central notions in the Ambient Intelligence vision to transform our living and working environments into ’
’. Ontology-based models have been argued to satisfy all demands concerning context representation. Rule-based reasoning has already been successfully integrated in ontology-based applications for domains with similar requirements (e.g. the Web), while it offers significant advantages concerning its deployment in the Ambient Intelligence domain. In this paper, we analyze the general challenges of contextual reasoning, argue about the suitability of rule-based reasoning, and describe the deployment of such methods in two different settings; in a centralized semantics-based context management framework for Ambient Intelligence, and in a totally distributed system of logic-based abient agents.